effects of the functional analysis celeration chart …
TRANSCRIPT
The Pennsylvania State University
The Graduate School
College of Education
EFFECTS OF THE FUNCTIONAL ANALYSIS CELERATION CHART ON DECISION
MAKING FOR BOARD CERTIFIED BEHAVIOR ANALYSTS
A Dissertation in
Special Education
by
Salvador Ruiz
Ó 2018 Salvador Ruiz
Submitted in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
May 2018
ii
The dissertation of Salvador Ruiz was reviewed and approved* by the following:
Richard M. Kubina Jr. Professor of Special Education Dissertation Adviser Chair of Committee Charles A. Hughes Professor of Special Education Jonte C. Taylor Assistant Professor of Special Education David Guthrie Associate Professor of Education
Mary C. Scheeler
Associate Professor of Special Education
Professor in Charge of Special Education
*Signatures are on file in the Graduate School
iii
ABSTRACT
Functional Analysis (FA) has contributed significantly to the field of Applied Behavior
Analysis. The use of visual analysis to determine function has largely proven effective.
However, challenges still exist, namely how can behavior analysts unify analysis and
interpretation of results. The present study will examine the extent to which Board Certified
Behavior Analysts (BCBA) could determine function of challenging behavior by using a
standardized graphical display called the Functional Analysis Celeration Chart (FACC).
Participants with BCBA certification received an invitation to participate in the survey.
Participant demographic questions, FA data recharted on the FACC, and decision making
questions will appear on visual displays and Likert scale questions.
Keywords: functional analysis, visual analysis, and functional analysis chart
iv
TABLE OF CONTENTS
LIST OF TABLES ..............................................................................................................vi
LIST OF FIGURES .............................................................................................................vii
ACKNOWLEDGEMENTS .................................................................................................viii
Chapter 1 INTRODUCTION ..............................................................................................1
Visual Analysis ............................................................................................................2 Supplements to Visual Analysis ...................................................................................3Present Study ...............................................................................................................4
Chapter 2 METHOD ..........................................................................................................6
Participants ..................................................................................................................6 Survey .........................................................................................................................6 Functional Analysis Celeration Chart ...........................................................................7 Determining Function ..................................................................................................7 Dependent Measures ....................................................................................................8 Procedure.....................................................................................................................8 Tutorial ........................................................................................................................9
Accuracy………………………………………………………………………………….... 9 Social Validity .............................................................................................................9
Chapter 3 RESULTS ..........................................................................................................11
Determining Function with the FACC ..........................................................................11 Quantification of Level ................................................................................................12 Preference on Graphical Display ..................................................................................13
Chapter 4 DISCUSSION ....................................................................................................14
Determining Function with the FACC ..........................................................................14 Quantification, Technology, and the FACC ..................................................................18 Preference on Graphical Display ..................................................................................18 Limitations ..................................................................................................................19 Future Directions .........................................................................................................20
References...........................................................................................................................21
Appendix A Survey ............................................................................................................25
Appendix B Tables .............................................................................................................50
Appendix C Figures ............................................................................................................52
v
Appendix D Literature Review ...........................................................................................55
Appendix E Recruitment Materials…………………………………………………………...95 Recruitment Email sent to the Behavior Analytsis Certificaiton Board ..........................95
Informed Consent………………………………………………………………………..96
vi
LIST OF TABLES
Table 1: Preference on the Sequential View FACC. .............................................................50
Table 2: Preference on the Nonsequential View FACC. ......................................................50
Table 3: Percentage of displays when Quantification was used in Decision Making .............51
Table 4: Participants Preference on Graphical Displays. .......................................................51
vii
LIST OF FIGURES
Figure 1: Participant Years of Experience Practicing Behavior Analysis. .............................52
Figure 2: Participant Years of Experience using Visual Analysis..........................................53
Figure 3: Participant Years of Experience Conducting/Analyzing Functional Analysis .........53
Figure 4: Correct Responding by type of Result. ..................................................................54
Figure 5: Correct Responding by Type of FACC Display. ....................................................54
viii
ACKNOWLEDGEMENTS
Completing a dissertation is a contribution to science. I can say that without the support of my committee, family, and friends I would not have had the ability to accomplish my goals. I wish to offer a wholehearted thank you to the following people. First, I would like to thank my mentor, Dr. Richard Kubina. The lessons you have taught me will serve me well moving forward and I feel extremely prepared, confident, and happy in all of my choices. I can’t express my appreciation for all you have done. Next, I would like to thank Dr. Charles Hughes. We have had many tough conversations and your guidance became invaluable as I navigated through the program. Dr. Taylor, thank you for your support, feedback, and guidance. I learned a lot from you. Dr. Guthrie, learning from you both in and out of the classroom was a great experience that I will never forget. Throughout my time at Penn State I received a lot of support from Dr. Wolfe and Dr. Lee. I thank both of you for your time and kind words when they were needed. I consider myself fortunate to receive so much encouragement from the Chartlytics team as well. Thank you for that. Last, I would like to thank my mother and father for all their support. I appreciate all that you have done and love both of you. To Tony and Megan, thank you so much for your great friendship when things haven’t been the best. I consider both of you family and your support helped throughout this process. Thank you to my friends who have provided words of encouragement and celebrations!
1
Chapter 1
Introduction
Functional Analysis (FA) provides guidance for the treatment of challenging behavior by
determining what type of reinforcement will maintain a target behavior. FA has several defining
features: (1) The assessment provides a series of test conditions and a control condition; (2) a
condition specific motivating operation and reinforcement for each test condition creates
deprivation and increases the likelihood of the occurrence of the target behavior; and (3) a control
condition that allows for a comparison against the test conditions (Neef & Peterson, 2007).
Recorded instances of the target behavior appear on a visual display for analysis. The level of
each data set determines the condition(s) that produced the most occurrences of the target
behavior when compared to the control.
FA has benefitted from widespread literature demonstrating the ability of the assessment
to determine function of challenging behavior, effectiveness of variations, training professionals
on implementation, development of sub-types of functions, the use of structured criteria to
determine agreement of function, and the use of statistical methods to augment visual analysis
(e.g. Brossart, Parker, Olson, Mahadevan, 2006; Journal of Applied Behavior Analysis, 1994;
JABA, 2003; JABA, 2013). Comprehensive reviews of functional analysis findings have further
demonstrated the effectiveness and importance of the issue (Hanley, Iwata, & McCord, 2003;
Beavers, Iwata, & Lerman, 2013). The use of FA in the treatment of problem behavior has led to
the development of function-based interventions and decreases in the use of punishment-based
interventions (Pelios, Morren, Tesch, & Axelrod, 1999).
Variations of FA have included slight modification to procedures and measurement
tactics. For example, Trial Based Functional Analysis modifies procedures by conducting the
assessment in the natural environment and will typically use a bar graph (See Rispoli, Ninci,
Neely, & Zaini, 2014 for a comprehensive review). The use of bar graph provides a direct view
2
of level, but discounts sequencing effects. Sequencing effects demonstrate if a participant has
learned to access reinforcement in an environment, which may lead to false positive results
(Johnston & Pennypacker, 2009). Additionally, if two or more conditions have a similar level of
behavior, the ability to correctly determine function may become difficult due to the inability to
see variability and trend on a bar graph.
Visual Analysis
The use of visual analysis has contradictory perspectives in the literature (e.g.
DeProspero & Cohen, 1979; Khang, Chung, Gutshall, Pitts, Kao, & Girolami, 2010). Research
has shown inconsistent analytical conclusions across analysts (Jones, Weinrott, & Vaugt, 1978;
DeProspero & Cohen, 1979; Ninci, Vannest, Wilson, & Zhang, 2015). Experimenter’s have
shown low agreement between statistical methods and visual analysis (DeProspero & Cohen,
1979; Jones et al., 1978). The findings suggest visual analysis may not provide enough evidence
to support effective interventions. For example, inaccurate analysis may occur due to an error in
the visual display. Further, the analyst may lack the ability to appropriately analyze the data. The
use of statistical methods provides additional checks such as confidence intervals and effect sizes
(Brossart, Parker, Olson, & Mahadevan, 2006; Michaels, 1974).
Additionally, research has shown that in Trial Based Functional Analysis conflicting
results may occur using different graphical displays with the same analysts (Ruiz & Kubina,
2017). The issue of reliability of analysis becomes imperative with regards to FA because of the
severity of the challenging behavior targeted. Notably, Bloom and colleagues did not match three
Trial Based FA’s to traditional FA and had one partial match (Bloom, Iwata, Fritz, Roscoe, &
Carreau, 2011). Similarly two other manuscripts either had a partial match or no match when
comparing FA and Trial Based FA (LaRue, Lenard, Weiss, Bamond, Palmieri, & Kelly, 2010;
Rispoli, Davis, Godwyn, & Camargo, 2013). The previously described findings indicate using
different graphical displays may produce different conclusions in regards to function.
3
Researchers have attempted to address visual analysis inconsistencies through two
methods. First, the use of structured criteria to determine the function of challenging behavior
with ten data points per condition and varying data points per condition (Hagopian, Fisher,
Thompson, Owen-DeSchryver, Iwata, & Wacker, 1997; Roane, Fisher, Kelley, Mevers, &
Bouxsein, 2013). Second, statistical methods that enhance the use of visual analysis to determine
if a functional relation exist (Davis, Gangnè, Fredrick, Alberto, Waugh, & Haardörfer, 2013).
Both methods have demonstrated effectiveness.
Supplements to Visual Analysis
The use of structured criteria provides standardized decision rules for professionals.
Using a criterion that provides rules facilitates the consistent examination of data for determining
a function. Development of structured criteria relies on an appropriately constructed display.
However, line graphs appear with variable construction features and with widely discrepant
proportions and scaling. For example, the axes often do not follow the proportional construction
rule, which governs the ratio of size between the vertical (5/8 to 3/4) to the horizontal axis
(Kubina, Kostewicz, Brennan, & King, 2017). An ill-formed graph visually affects the placement
of trend, variability, and level.
To mitigate the many subjective properties of nonstandard linear graphs (e.g. scaling,
labeling, placement of tick marks), researchers have experimented with statistical methods and
begun pairing statistical analysis with visual analysis (Brossart et al., 2006; Davis et al., 2013;
Parker, Cryer, & Byrns, 2006). The use of quantification can provide a level of certainty
supporting visual analysis. For example, FA data sets with multiply maintained results can have
elevated levels of behavior in several test conditions. The test conditions may not have an equal
level to each other and outliers may cloud judgment.
Providing a numerical value for the occurrence of behavior in each test condition can
facilitate the identification of interventions most effectively leading to a decrease in targeted
behavior. While visual analysis will show elevated levels of behavior in a test condition, the
4
variability of the data path may cause concern in determining if a test condition maintains a
behavior. Quantification augments visual analysis by using a measure of central tendency across
a condition to determine the value of the level. Also, quantification of variability reflects the
degree of control or influence of the present condition on the target behavior.
Another possibility aside from supplementing visual analysis with statistical analysis
resides in having a chart that directly offers statistical information - the Standard Celeration Chart
or SCC (Pennypacker, Gutierrez, & Lindsley, 2003). The SCC, a standardized ratio chart, has
graphic properties allowing the chart reader to not only visualize the data in a standard capacity,
but also quantify the change features (e.g., trend – celeration, variability – bounce). The
numerical values improve visual analysis, and direct statistical information on the occurrence of a
target behavior during a test condition.
Present Study
A modified version of the SCC called the Functional Analysis Celeration Chart (FACC)
can quantify level and visually analyze FA data sets (See Figure 1 and 2). The FACC displays
data in a sequential and condition-grouped view, offering a numerical comparison of level across
conditions, and uses different symbols to represent each condition (Chartlytics, 2016). The
different data displays each provide a unique advantage. The sequential display allows for
detection of sequencing effects, which assist in determining a potential problem with procedures.
For example, a participant learning how to access reinforcement in the session would have an
increase in behavior across time. The condition-grouped display affords an easier visual
inspection of the data by grouping each condition on the graph (See Figure 2). When an
undifferentiated result occurs, experimenter’s will typically conduct additional sessions testing
one function at a time called an extended analysis, or pairwise analysis (Betz & Fisher, 2011).
The additional sessions expose the participant to continued reinforcement, potentially
strengthening the response. Additionally, the FACC delivers numerical values that eliminate the
need for pairwise analysis assuming behavior occurs in any of the conditions.
5
The FACC may further assist with undifferentiated results. Undifferentiated results occur
when the levels of problem behavior are variable across conditions, without any of the test
conditions displaying a higher pattern of responding than the control (Betz & Fisher, 2011).
Current practice suggests extending the analysis to conduct additional sessions if the behavior
remains undifferentiated a retest of socially mediated conditions may provide conclusive results
(Betz & Fisher, 2011). However, the FACC quantifies existing data contributing guidance for
which condition yielded the most behavior. While the pattern may appear undifferentiated with
visual analysis alone, the use of quantification as an analytical tactic allows for the development
of an intervention.
The proposed study seeks to examine the effectiveness of the FACC by measuring
participants’ ability to detect the function of challenging behavior in sequential and conditioned
grouped views. Further, determining the role of quantification and preference may predict use of
the display moving forward. Specifically, the present study asks the following questions:
1. To what extent will Board Certified Behavior Analysts determine the function of
challenging behavior using the FACC?
2. How frequently will BCBA’s use the quantification of level to determine the function of
challenging behavior?
3. Will participants find a standardized graphical display preferable for analysis in
comparison to equal interval line graphs?
6
Chapter 2
Method
Participants
The participants had a BCBA certification and received a request for participation via the
Behavior Analyst Certification Board’s website. All participants completed coursework in
Applied Behavior Analysis, a minimum of 1,500 experience points, and passed a certification
exam. Additional questions inquired about years of experience practicing behavior analysis, time
spent interacting with graphical displays of behavior, and experience conducting FA assessments.
Further, questions pertained to type of employment held to clarify roles of participants.
Participants had the option of not taking part in the survey. Anyone who began the survey but did
not complete had their answers excluded from the results.
Participant Characteristics. A total of 20,956 people received a request to take the
survey. All participants have a BCBA credential and have their contact information on the
Behavior Analysis Certification Board Registry (BACB.com). Behavior Analysts that begin the
survey can enter their job responsibilities in a provided text box. Job responsibilities may include
direct service implementation, consulting with families whose children receive behavior analytic
services (in school, or home), and provide consultation to school districts for children that qualify
for behavior analytic services. All participants have a minimum of 15 credit hours in behavior
analytic content and completed a minimum of a master’s degree.
Survey
A survey administered via Qualtricsâ asked participant demographic questions, as well
as, display charts with functional analysis data (See Appendix A). All FA data selected has
appeared in published manuscripts and recharted on a Functional Analysis Celeration Chart by
the experimenter. The selected FA data will come from a highly referenced study (Iwata, Dorsey,
Slifer, Bauman, & Richman 1994). The function for each target behavior appears in the original
7
manuscript. Additional FA data came from an article that implemented visual criteria for
decision making rules in-regards to determining the function of a challenging behavior (Roane,
Fisher, Kelley, Mevers, & Bouxsein, 2013). The manuscripts demonstrated functional relations
of behavior in the selected graphs. Manuscripts selected had used traditional FA procedures.
After the participants determine the function of each graph, they would answer additional
questions regarding the method they used to determine function and preferences of analytical
tactics.
Functional Analysis Celeration Chart (FACC)
The online Chartlytics graphing platform allows for quantification of data and a standard
visual display. The FACC displays functional analysis data using a standard, ratio line chart
(Chartlytics, 2016). The data display can appear in a sequential, or non-sequential (i.e. grouped
by condition) order. Standardized symbols represent each main condition and the user sets what
the symbols represents. For example, an open circle can symbolize an alone condition and a star
can depict a demand condition. The FACC also quantifies data, (e.g. level). A user who hovers a
cursor over the horizontal level line sees the numerical value of the level. Each condition has
level values appearing in a table facilitating a quantifiable comparison for all conditions
conducted in an FA. Software users can select different options for level calculation: geometric
mean, arithmetic mean, or median. FACCs included in the survey will have a static display
generated by screen shots of data. The displays contained data with level lines and a table with
the level values (See Figures 1 & 2).
Determining Function
The survey asked participants to select from condition specific answers, a combination of
conditions (e.g. attention, escape, tangible), or undifferentiated results. Each data set had the
possibility of behavior maintained by one function, multiple functions, or undifferentiated (i.e.,
unable to determine a function). In all, 5-6 choices will appear under each data set. If a data set
demonstrated behavior maintained by multiple functions, participants had to select each function.
8
If a participant does not select all of the functions in a multiply maintained result, the selection
scores as an incorrect response. Participant responses, recorded electronically, receive a score of
correct (response matched the function in the manuscript) or incorrect (responses did not match
the function in the manuscript).
Dependent Measures
The dependent variable consisted of FA data from published manuscripts, recharted on an
FACC. Each FACC survey question has the ability to select functions that could maintain the
behavior (See Figure 3). Of the 18 graphs included in the survey, nine charts have a one-function
result (n=9), two charts had an undifferentiated result, and six (n=6) have a multiply maintained
result. Demographic and analysis questions rely on a Likert scale to quantify participant
responses (See Figure 4 & 5).
Procedure
A recruitment email with a link to the survey sent to participants requested participation.
In addition, the email provided a rationale for the study. Prior to accessing the survey,
participants received a tutorial on visual analysis with the FACC. The tutorial included the
components of an FACC in both views (Sequential and Condition-Grouped), FA data,
quantification table, and how to analyze data using the FACC. Participants selected the survey
via an embedded link. The email contained information regarding identification of selection, a
link to the survey, and information regarding a chance to win a $50.00 Visa Gift Card. The first
page included demographic questions. Each subsequent page contained one FACC per page and
a table with the numerical values with the option of selecting if the data show one function (with
each function listed as an option), multiply maintained, or undifferentiated results. Participants
could select more than one option per question. The last page included participants reacting to
the FACC and decision making tactics.
9
Tutorial
A pre-recorded tutorial explained the use of visual analysis (i.e. level as a decision
making tactic) and how to read the quantification of the level (i.e., calculated with the geometric
mean) on the FACC lasted approximately 10 minutes. Data appeared in both sequential and non-
sequential views. The tutorial included two types of visual displays. A sequential view FACC
shows data visually and with a table that compare each condition to the control. Participants
received instruction on how the sequential view FACC displays data and how to read the table to
analyze the data. A condition-grouped FACC displayed the same data as the sequential view
FACC with instruction on the differences between the two types and examining the level of each
condition.
Accuracy
Minimizing error in measurement allows experimenters the ability to make better
decisions. A true value represents a standard value arrived at through extra efforts involved in
minimizing measurement error. (Kostewicz, King, Datchuk, Brennan, & Casey, 2016). The use
of accuracy portrays how closely observed values approximate what actually happened (Johnston
& Pennypacker, 2009). The present study established a true value by examining the participant
responses (i.e., the observed values) on each survey in relation to the responses provided by the
participants through Qualtrics®. Participants could perform the behavior by selecting a response.
An instance of a response demonstrates an instance of behavior (i.e. participants selected a
response from the options available). The experimenter viewed selected responses to each survey
question and compare the responses to the results from Qualtrics® (i.e., the true value).
Social Validity
Social Validity guides program planning and evaluation (Schwartz & Baer, 1991). Upon
completion of participant responses to how analysis and function judgments occurred, additional
questions asking about the preference of using the FACC followed. The Likert scale included a
range from strongly disagree to strongly agree. Strongly disagree would indicate that the FACC
10
would not have preference as a possible graphical display for measuring FA data and strongly
agree would indicate that the display had preference.
11
Chapter 3
Results
A total of 503 participants responded to the survey and answered a minimum of one
question. Not all participants responded to every question. Further, a link for a tutorial included
in the participation request allowed the experimenter to track number of views. A total of 53%
participants watched the tutorial. All participants had obtained a BCBA™ prior to participation
in the study. Demographic questions indicated 35.17% of the sample population had 4-7 years of
experience practicing behavior analysis (See Figure 1). Additionally, nearly 40% had 4-7 years
practicing visual analysis (See Figure 2) indicating a disconnect between year’s experience and
examining graphical displays. The majority of respondents had 0-3 year’s experience conducting
and/or analyzing FA data (See Figure 3). Demographic results suggest participants had more
experience with practicing behavior analysis and visual analysis, with less exposure to FA.
Determining Function using the FACC
A total of nine sets of data appearing in sequential and condition grouped views yielded a
total of 18 graphs. Of the visual displays included in the current study, 10 had a single function
result, four showed multiply maintained functions, and four had no differentiation in responding.
The FA literature demonstrates a large portion of single function results, followed by multiply
maintained (Beavers et al., 2013). Across the three types of results, the sample responded to
single function results most accurately (See Figure 4).
Single Function Results
All of the common functions of behavior had multiple displays (e.g. Attention, Demand,
Tangible, and Automatic). The automatic function choice on each question had a label of alone
to depict the test condition label in the graphical display. Each participant had the option of
selecting multiple response for every question regarding graphical displays. Correct responding
for single function results occurred 62.47% of the total responses.
12
Multiply maintained results. Multiply maintained results occur when the control
condition remains stable with a low level and more than one test has an elevated level of behavior
(Betz & Fischer, 2011). Two graphs had two functions maintaining the target response. An
additional two graphs had three functions maintaining the behavior. These survey questions
required participants to select all the test conditions that had an elevated level of behavior in
comparison to the control. The sample of participants responded correctly 31.7% of the total
responses (See Figure 4).
Undifferentiated results. Undifferentiated patterns of responding occur when no
discernable difference in level between test and control take place. Further, the level of any of
the conditions should place at the bottom of the graphical display (Betz & Fischer, 2011). A
raised level of behavior in each session would likely indicate an automatic function. A total of
four questions in the two views had a correct response of undifferentiated. The response choice
appeared in all (n= 18) display questions. Participants correctly responded to 49.8% of the total
number of opportunities (See Figure 4).
Response by type of display. Each display had the same data set presented in both
views. For example, if one data set appeared in a sequential view it similarly appeared in a
condition grouped view. Participants responded correctly to approximately 53% of the visual
displays in either view (See Figure 5), indicating that neither display evoked significantly more
correct responding. However, in the condition grouped view an outlier appeared. One FACC had
three test conditions maintaining the behavior. Participants largely selected undifferentiated
pattern of responding (38.4%). If calculating the average correct responding with the median,
condition grouped correct responses occurred for 58% of the total questions.
Quantification of Level
Level lines on visual displays use measures of central tendency to determine line
placement. The FACC used the Geometric Mean to determine the level line. The value of each
conditions level can then receive comparison to any other condition. For example, if the control
13
condition shows no change (i.e. no change in responding) the level would then show x1, any
condition compared to the control with an elevated level of responding would display at a higher
value indicating more responding occurred. Participants had access to the level values for each of
the 18 displays. When asked how frequently participants used the table for decision making
34.8% of respondents reported examining the table for 0-5 displays reflecting a heavy reliance on
visual analysis (See Table 3). However, 28% of the sample reported using the table for most of
the FACC charts.
Preference on Graphical Display
Participants reported on preference of graphical display and satisfaction with each display
(i.e. sequential vs. condition grouped). When asked to report on preference of graphical display a
visual image of a line graph received inclusion although no survey question included the image.
Most respondents reported their greatest preference for the line graph (See Table 4).
Additionally, 3.2% of the sample reported no preference for any of the three options. A Likert
scale ranging from strongly satisfied to strongly dissatisfied appeared to determine preference of
the two FACC displays. Nearly half of the sample reported some level of satisfaction with the
sequential FACC (See Table 1). Similar results occurred for the condition grouped view.
Preference on display did not vary based on display.
14
Chapter 4
Discussion
A review of the relevant literature indicates that visual analysis requires support to
maximize effectiveness (See Appendix D). As an assessment, functional analysis or FA uses the
graphical element “level” as the primary analytic tactic. Previous studies have shown a low level
of agreement when using only visual analysis sans intervention (e.g. structured criteria; visual
aids). The use of structured criteria has demonstrated an increase in agreement across raters,
increasing the likelihood of correct identification of function (Hagopian et al., 1997; Roane et al.,
2013). However, the criteria do have limitations. Primarily, the criteria rely on a truth-by-
agreement standard. While agreement may provide more confidence in analysis, consensus
among observers does not provide any information on the accuracy of the observed events
(Johnston & Pennypacker, 2009; Kostewicz et al., 2016). The Functional Analysis Celeration
Chart (FACC) can help behavior analysts with function detection through objective
quantification. The present study asked the following questions:
1. To what extent will Board Certified Behavior Analysts determine the function of
challenging behavior using the FACC?
2. How frequently will BCBA’s use the quantification of level to determine the function of
challenging behavior?
3. What preference will participants have for a standardized graphical display?
Determining Function using the FACC
Participants viewed data on the FACC and had an opportunity to select the correct
function(s) maintaining the behavior in each visual display. Further, participants could choose
multiple responses for each display question. The results showed correct responding occurred
most frequently with single function results, followed by undifferentiated, and then multiply
maintained.
15
Single Function Results. Most FA results that appear in the literature have a single
function maintaining the response (Beavers et al., 2013). That makes the ability to identify single
function results important for practicing behavior analysts because of the likelihood of
encountering one during analysis. Determining a single test condition maintaining a behavior
relies on low levels of the control and the other conditions. Visually a clear separation in level
must occur.
The current study found single function results as the most commonly identified correct
function suggesting that the FACC allows analysts to clearly determine one function of a
challenging behavior. A similar study found 46% interrater agreement across analysts (Hagopian
et al., 1997). However, the authors did not specify agreement across types of results.
Comparatively, participants in the current study performed at higher levels (i.e., 62% function
identification). Further, the data used had come from well referenced articles that identified a
function. The participants determined the same function as the previously referenced studies on
the FACC.
The implications for higher identification scores in the present study suggest behavior
analysts can use a standard ratio graph to determine function. The FACC has a ratio scale which
offers ratio comparisons of data sets with different magnitudes, show growth or rate of change
better than linear graphs, and normalize variance (Few, 2009; Harris, 1999; Kubina & Yurich,
2012). The view of the data on the ratio scale may have led to cleaner comparisons between
different test conditions. Additionally, the variability may have also flattened thereby showing
more stability and thus a clear function.
Multiply Maintained Results. Multiply maintained results can occur when multiple test
conditions have an elevated level in comparison to the control (Betz & Fischer, 2011).
Participants had varying degrees of correct responding. Specifically, the FACC charts with two
functions maintaining the challenging behavior resulted in participants answering correctly more
often than when a behavior had three conditions maintaining the response. When multiple test
16
conditions maintain a behavior, detecting function proves challenging. For example, the level of
the control condition would dictate if a behavior is maintained by other tests or automatic
reinforcement. Visual analytic tactics (e.g. trend, stability, variability) have shown to create
issues with FA results (Diller, Barry, & Gelino, 2016). It remains unclear if the ratio scale of the
FACC helped or hindered the process of function identification.
The FACC does advance the use of quantification of the levels, a seldom-used approach
for determining function. Quantification has proven effective with other types of research
designs (e.g. ABAB) and using statistical testing (Brossart et al., 2006; Davis et al., 2013). The
current study differs in that it focused on multielement designs. The effects of quantification did
not seem apparent as participants did not respond correctly frequently. A tutorial provided
explained how to use quantification. However, half of the sample did not view the tutorial that
reviewed the quantification of level. The tutorial would have demonstrated how quantification
could assist when determining multiply maintained results. With only half of the sample
watching the tutorial, determining the effect of quantification of level with the table requires
future study.
Undifferentiated Results. Undifferentiated results occur in approximately 5% of the FA
literature (Beavers et al., 2013; Hanley et al., 2003) indicating that behavior analysts cannot
typically identify functions of behavior. The current study had a small sample of undifferentiated
FA results as well. An undifferentiated result occurs when all conditions (e.g. tests, control)
provide no visually discernable difference in level (Betz & Fischer, 2011). Essentially the analyst
looks for a pattern in responding and when one does not seem apparent an FA has
undifferentiated result. It remains unclear how quantification or visual analysis of the FACC
impacted undifferentiated results.
Participants detected an undifferentiated result using the FACC for half of the graphs.
Over half the sample correctly identified undifferentiated responses in two separate questions.
Each question had a different view (i.e., sequential and condition grouped). Quantification may
17
have impacted decision making as the level values did not have numerical separation. Visually,
the non-sequential view demonstrates similar level lines across conditions and sequentially the
data do not separate and remain at low levels. While research in the reliability of detecting
functions on multi-element designs appears limited, one study found low levels of reliability
across analysts (Diller et al., 2016). The researchers discovered level distances created issues
with analysis. The issue appeared most prominently for variability within conditions. Raters
performed poorly when data paths have variability and did not reach agreement. The participants
in the current study may have experienced a similar issue with undifferentiated and multiply
maintained responding.
Response by type of display. The FACC has the unique ability to examine data in
sequential and condition grouped views. To date, no other study has examined a condition
grouped view of data on a ratio scaled visual display. The condition grouped view takes the data
out of time, but allows for easy viewing of the level of each condition. The sequential view
places the data in the order the conditions transpired, allowing for a more traditional view of data.
Participants performed slightly better in the condition grouped view (See Figure 5) suggesting
that grouping by condition may provide better analysis. The grouped data may more distinctly
present differences between the conditions when compared the data laid on top of each.
The condition grouped FACC collates data by condition which also differs from a bar
graph in that analysts can see all sessions. The ability to scan and see where the level lines lay in
each condition make decision making efficient. Trial Based Function Analysis (TBFA) uses bar
graphs in many instances, however, the data do not appear individually (See Ruiz & Kubina,
2017 for a comprehensive review). Condition grouping while still observing individual data
points appears exclusively on the FACC.
Despite overall higher rates of correct responding, an outlier appeared in the condition
grouped FACC. Many participants responded to one condition grouped view incorrectly leading
to a 5% decrease in overall correct responses. When using the median, the correct response rate
18
increased when compared to the sequential view. The outlier had three functions maintaining the
behavior. If participants did not select all three function responses, the question scored incorrect.
While the outlier only appeared in one question, more research examining multiply maintained
FACCs in the condition grouped view would provide more clarity.
Quantification, Technology, and the FACC
As stated earlier, quantification can assist with decision making (Hagopian, Rooker, &
Zarcone, 2015). The FACC calculates level line values and allows comparison with every
condition. The level offers an individual performance statistic for each condition stating the
average rate of responding immediately available to individual analysts. However, as stated
earlier it seems participants largely did not use the table as nearly half of the sample did not even
view the tutorial. The explanation of how to read the table appeared in the tutorial as a formal
decision making tactic for determining function. Therefore, the present results do not reflect the
potential for behavior analysts to use quantification for analysis and function detection.
The present survey questions used static displays, live technology and an understanding
of the data table and subsequent statistics could prove useful in the future. A recent study
demonstrated how telehealth technology allowed behavior analysts 200 miles away from clients
engage in successful functional analyses (Lee et al., 2015). The technology of conducting
functional analyses remotely still has many issues that need addressed, but in the present study
behavior analysts did better than average with single function results. Therefore, the FACC as
technology tool holds promise for displaying data for FAs.
Preferences for Graphical Displays
Within contemporary behavior analysis BCBAs have the option to choose whatever type
of visual display they like. Currently, the field of behavior analysis exclusively uses line graphs
to display, analyze, and communicate data (Cooper et al., 2007). A question asked participants
preferences for visual displays. While most participants preferred the line graph, the sample did
report satisfaction with the FACC. Over half the sample population reported satisfaction with the
19
sequential FACC and half with the non-sequential view (See Table 1). Willingness to accept and
use the FACC may help in future studies.
The benefits of the FACC have common features with other standardized ratio graphs.
For example, the Standard Celeration Chart (SCC) allows for quick and reliable interpretation of
data (Datchuk & Kubina, 2011). Further, the SCC quantifies data by using celeration, bounce,
and level (Datchuk & Kubina, 2011). The FACC shares standardized construction and
quantification features with the SCC. Exposing participants to the display will allow for
consistency in analysis as each display follows similar features. Further, previous research that
examines agreement with multi element designs uses a standardized construction and
quantification feature for linear graphs (Hagopian et al., 2015). The use of both has increased
agreement. Similarly, both tactics appear on the FACC.
History and experience with linear graphs may have contributed to the strong preference
towards line graphs. Further, participants may not have had previous exposure to the FACC. A
recently published meta-analysis that examined multiple factors in-regards to agreement
determined that experience may play a role in visual analysis (Ninci et al., 2015). More exposure
and practice using the FACC may impact its overall preference among behavior analysts.
Limitations
The present study has several limitations. First, one of the survey questions resulted in a
large amount of incorrect responding indicating a potential issue with the item. An outlier in
either the test or control conditions may have caused an error in responding. Outliers have
impacted decision making of analysts (Nelson, Van Norman, & Christ, 2016) making the need for
quantifiable values of level across a data set in FA important. The response had to include three
functions maintaining the behavior. However, each survey question had the same wording.
Second, although quantification of the level of each condition appeared on each graph a decision
making rule did not. For example, a difference in level of x2 would indicate function. A rule
may have increased correct responding. Third, more participant responses could provide more
20
information on the usefulness of the FACC as a tool. Fourth, participants did not have explicit
knowledge that multiple choices could receive selection per question. Regardless of the amount
of conditions that could maintain the behavior, every question could have all options selected.
Future Directions
Future research should provide increased response rates, as well as, the development of a
numerical value that would indicate function to support visual analysis. The value should receive
calculation by relying on published studies that included a determined function. Additionally, the
FACC needs more research to better establish its utility. From basic components such as
examining levels in each condition to viewing trends and variability in the data, future research
would show the technical merits of the FACC. Using the chart with variations of traditional FA
will provide additional insight and analysis.
Conclusion
The FACC has the potential to assist in determining functions of challenging behavior.
Its standardized construction features and quantification of level can befit users. Participants in
the current study correctly identified function most frequently with single function results
followed by undifferentiated results. With the development of a rule, quantification can assist in
decision making and future research should seek to increase response rates.
Correct responses exceeded previous research baseline levels (Hagopian et al., 1997).
Further, providing another graphical display to assist analysts in circumstances where other
displays may no demonstrate a functional relation. Visual Analysis with supports has been
proven to assist those seeking to determine relations with intervention and behavior (Davis et al.,
2013; Nelson et al., 2016; Ninci et al., 2015). The FACC provides those supports.
21
References Beavers, G. A., Iwata, B. A., & Lerman, D. C. (2013). Thirty years of research on the
functional analysis of problem behavior. Journal of Applied Behavior Analysis, 46, 1-21.
doi: 10.1002/jaba.30
Betz, A. M., & Fischer, W. W. (2011). Functional Analysis history and methods. In
Fisher, W. W., Piazza, C. C., & Roane, H. S., (Eds.). Handbook of Applied Behavior
Analysis. New York: Guildford Press.
Brossart, D. F., Parker, R. I., Olson, E. A., & Mahadevan, L. (2006). Relationship
between visual analysis and five statistical analyses in a simple AB single-case research
design. Behavior Modificaiton, 30, 531-563. doi: 10.11.77/0145445503261167
Cooper, J., O., Heron, E., T., Heward, W. L. (2007). Applied Behavior Analysis (2nd ed.).
Upper Saddle River, N.J.: Pearson Prentice Hall.
Datchuk, S., M., & Kubina, R., M. (2011). Communicating in experimental finding in
single case research: How to use celeration values and celeration multipliers to measure
direction, magnitude, and change. Journal of Precision Teaching and Celeration, 27, 3-
17.
Davis, D. H., Gagnè, P., Fredrick, L. D., Alberto, P. A., Waugh, R. E., & Haardörfer, R.
(2013). Augmenting visual analysis in single-case research with hierarchical linear
modeling. Behavior Modification, 37, 62-89. doi: 10.1177/0145445512453734
DePropsero, A., & Cohen, S. (1979). Inconsistent visual analyses of intrasubject data.
Journal of Applied Behavior Analysis, 12, 573-579. doi: 10.1901/jaba.1979.12-573
Diller, J. W., Barry, R. J., & Gelino, B., W. (2016). Visual analysis of data in a multielement
design. Journal of Applied Behavior Analysis, 49, 980-985. doi: 10.1002/jaba.325
Few, S. (2009). Now you see it: Simple visualization techniques for quantitative analysis.
Oakland, CA: Analytics Press.
22
Hagopian, L. P., Fisher, W. W., Thompson, R. H., Owen-DeSchryver, J., Iwata, B. A., &
Wacker, D. P. (1997). Toward the development of structured criteria for interpretation of
functional analysis data. Journal of Applied Behavior Analysis, 30, 313-326.
doi: 10.1901/jaba.1997.30-313
Hagopian, L. P., Rooker, G., W., & Zarcone, J., R. (2015). Delineating subtypes of self-injurious
behavior maintained by automatic reinforcement. Journal of Applied Behavior Analysis,
48, 523-543. doi: 10.1002/jaba.236
Hanley, G. P., Iwata, B. A., & McCord, B. E. (2003). Functional analysis of problem
behavior: A review. Journal of Applied Behavior Analysis, 36, 147-185.
doi:10.1901/jaba.2003.
Harris, R. L. (1999). Information graphics: a comprehensive illustrated reference. New
York: Oxford University Press.
Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & Richman, G. S. (1994).
Toward a functional analysis of self-injury. Journal of Applied Behavior Analysis, 27,
197-209. doi:10.1901/jaba.1994.27.197
Johnston, J. M., Pennypacker, H. S., & Green, G. (2009). Strategies and tactics of
behavioral research. New York, NY: Routledge.
Jones, R. R., Weinrott, M. R., & Vaught, R. S. (1978). Effects of serial dependency on
the agreement between visual and statistical inference. Journal of Applied Behavior
Analysis, 11, 277-283.
Kahng, S. W., Chung, K.-M., Gutshall, K., Pitts, S. C., Kao, J., & Girolami, K. (2010).
Consistent visual analyses of intrasubject data. Journal of Applied Behavior
Analysis, 43, 35–45. doi: 10.1901/jaba.2010.43-35
Kostewicz, D. E., King, S. A., Datchuk, S. M., Brennan, K. M., & Casey, S. D. (2016).
Data collection and measurement assessment in behavioral research: 1958-2013.
Behavior Analysis Research and Practice, 16, 19-33. doi: 10.1037/bar0000031
23
Kubina, R. M., Kostewicz, D. E., Brennan, K. M., & King, S. A. (2017). A Critical Review of
Line Graphs in Behavior Analytic Journals. Educational Psychology Review, 29, 583-
598. DOI 10.1007/s10648-015-9339-x
Kubina, R. M., & Yurich, K. K. L. (2012). The Precision Teaching Book. Lemont, PA:
Greatness Achieved.
LaRue, R. H., Lenard, K., Weiss, M. J., Bamond, M., Palmieri, M., & Kelley, M. E.
(2010). Comparison of traditional and trial-based methodologies for conducting
functional analyses. Research in Developmental Disabilities, 31, 480-487.
doi:10.1016/j.ridd.2009.10.020.
Lee., J. F., Schieltz, K., M., Suess, A., N., Wacker, D., P., Romani, P., W., Lindgren, S., D.,
Kopelman, T., G.,… Padilla Dalmau, Y. (2015). Guidelines for developing telehealth
services and troubleshooting problems with telehealth technology when coaching parents
to conduct functional analyses and functional communication training in their homes.
Behavior Analysis Practice, 8, 190-200. doi: 10.1007/s40617-014-0031-2
Michael, J. (1974). Statistical inference for individual organism research: Some reactions
to a suggestion by Gentile, Roden, & Klein. Journal of Applied Behavior Analysis, 7,
627-628.
Neef, N. A., & Peterson, S. M. (2007). Functional behavior assessment. In Cooper
J. O., Heron, W. L., & Heward, T. L. (Eds.). Applied behavior analysis,
(2nd ed). Upper Saddle River, NJ: Pearson Prentice Hall.
Nelson, P. M., Van Norman, E. R., & Christ, T. J. (2016). Visual Analysis Among
Novices: Training and Trend Lines as Graphic Aids. Contemporary School Psychology,
1-10. doi: 10.1007/s40688-016-0107-9
Ninci, J., Vannest, K. J., Willson, V., & Zhang, N. (2015). Interrater agreement between
visual analysts of single-case data: A meta analysis. Behavior Modification, 39,
510-541. doi: 10.1177/0145445515581327
24
Parker, R. I., Cryer, J., & Byrns, G. (2006). Controlling baseline trend in single-case
research. School Psychology Quarterly, 21, 418-443.
Pennypacker, H. S., Gutierrez, A., & Lindsley, O. R. (2003). Handbook of the standard
celeration chart. Cambridge, MA: Cambridge Center for Behavioral Studies.
Rispoli, M. J., Davis, H. S., Goodwyn, F. D., & Camargo, S. (2012). The use of trial-
based functional analysis in public school classrooms for two students with
developmental disabilities. Journal of Positive Behavior Interventions, 15, 180-189.
doi:10.1177/1098300712457420.
Rispoli, M., Ninci, J., Neely, L., & Zaini, S. (2014). A systematic review of trial-based
functional analysis of challenging behavior. Journal of Developmental Physical
Disability, 26, 271-283. doi: 10.1007/s10882-013-9363-z
Roane, H. S., Fisher, W. W., Kelley, M. E., Mevers, J. L., & Bouxsein, K. J. (2013).
Using modified visual-inspection criteria to interpret functional analysis outcomes.
Journal of Applied Behavior Analysis, 46, 130-146. doi: 10.1002/jaba.13
Ruiz, S., & Kubina, R. M. (2017). Impact of trial-based functional analysis on
challenging behavior and training: A review of the literature. Behavior Analysis Research
and Practice. doi: 10.1037/bar0000079
Schwartz, I. S., & Baer, D. M. (1991). Social validity assessments: Is current practice
state of the art?. Journal of Applied Behavior Analysis, 24(2), 189-204.
25
Appendix A
Survey
Figure 1: Sample FACC in Sequential View
26
Figure 2: Sample FACC in Condition Grouped View
27
Figure 3: Sample Question FACC Condition Grouped View
Figure 4: Sample Demographic Question
28
Figure 5: Sample Preference Question
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Appendix B
Tables
Table 1: Preference on the Sequential View FACC. Satisfaction with Sequential View FACC Percentage Reported
Strongly Satisfied
8.87%
Satisfied
25.81%
Somewhat Satisfied
17.74%
Neither Satisfied or Dissatisfied
20.97%
Somewhat Dissatisfied
12.1%
Dissatisfied
10.89%
Strongly Dissatisfied
3.63%
Table 2: Preference on the Nonsequential View FACC. Satisfaction with the Nonsequential View FACC Percentage Reported Strongly Satisfied
7.63%
Satisfied
22.49%
Somewhat Satisfied
19.28%
Neither Satisfied or Dissatisfied
23.29%
Somewhat Dissatisfied
12.05%
Dissatisfied
11.24%
Strongly Dissatisfied
4.02%
51
Table 3: Percentage of displays where quantification was used in decision making. Decision Making Using Quantification Percentage Reported
0-24% of Displays
34.8%
25-49% of Displays
16%
50-74% of Displays
21.2%
75-100% of Displays
28%
Table 4: Participant preferences on graphical displays.
Preference of Graphical Display Percentage Reported Sequential View FACC
12.4%
Nonsequential View FACC
10%
Equal Interval Line Graph
74.4%
None
3.2%
52
Appendix C
Figures
Figure 1: Participant years of experience.
53
Figure 2: Participant years of experience using visual analysis.
Figure 3: Participant years of experience conducting/analyzing FA assessments and data.
0
10
20
30
40
50
60
70
80
90
100
0-3 4-7 8-10 11+
Using Visual Analysis
0
10
20
30
40
50
60
70
80
90
100
0-3 4-7 8-10 11+
54
Figure 4: Correct responding by type of result on the FACC.
Figure 5: Correct responding by type of FACC display.
0
10
20
30
40
50
60
70
80
90
100
Single Function Multiply Maintained Undifferentiated
Type of Result
0
10
20
30
40
50
60
70
80
90
100
Sequential View FACC Condition Grouped FACC
55
Appendix D
Review of the Relevant Literature Outline
Functional Analyses (FA) have a successful history in the field of behavior analysis. Current
procedures and practices closely mirror work originally published in 1982 and reprinted in 1994
(Iwata, Dorsey, Slifer, Bauman, & Richman, 1994). Essentially, experimenters developed an
assessment that superimposed conditions with a specific reinforcer and collected information on
the occurrence of the target behavior. The visual data displayed would suggest a function of
behavior when a test condition occurred at a discernable difference in level higher than a control
condition (Iwata et al., 1994). The success of the assessment has led to the publication of three
special issues in the Journal of Applied Behavior Analysis, numerous variations of the procedure,
and wide spread adoption by behavior analysts (JABA 1994; 2003; 2013).
FA relies on visual analysis to determine function. Primarily, using a multielement
design, a variation of an alternating treatment design, FA compares each test condition to a
control condition (Cooper, Heron, & Heward, 2007). The reliability of visual analysis requires
raters to determine a functional relation between a dependent (e.g. behavior) and independent
variable (e.g. intervention/assessment). If an analyst has difficulties determining the effects of an
independent variable a need exists to improve it. In order to ensure raters can agree on what data
show, graphical displays must follow appropriate construction guidelines. In concert with such
rules, visual aids also help guide analysis of the data.
Implementing visual aids on graphical displays supports determination of the degree of
change in the data, the course of the data, and the level of the overall data set. Visual aids include
examples such as trend and level lines (e.g., Nelson, Van Norman, & Christ, 2016). Visual aids
should provide additional guidance to analysts. Nevertheless, construction issues may skew the
aids making data appear to have more dramatic effects. Pairing visual aids with statistical tests, a
56
suggestion by some researchers, may then create disagreement between visual analysis and
quantification.
Manuscripts examining the literature on FA have analyzed publication rates across
journals, participant characteristics, topographies of behavior, and methodological characteristics
(Beavers, Iwata, & Lerman, 2013; Hanely, Iwata, & McCord, 2003). The current review differs
in that it examines the reliability of visual analysis, how experimenter’s have made visual
analysis more reliable, and investigates the occurrence of errors in the construction of FA graphs.
More specifically, the current review asks the following questions:
1. Does visual analysis offer a reliable method to analyze behavior data?
2. What suggestions have experimenters made to improve visual analysis?
3. Do construction errors appear in the FA literature?
Reliability of Visual Analysis
Applied Behavior Analysis uses visual analytic tactics to determine if a functional
relation exists between interventions and targets behaviors (Cooper et al., 2007). FA differs in
that the assessment does not function as an intervention but instead reinforces instances of the
challenging behavior. After the assessment, the level of a test condition(s) will occur at higher
rates than the control (Betz & Fischer, 2011). However, the need to determine the differences in
level between test and control conditions relies on the visual display of data. Issues regarding the
reliability of visual analysis have received attention in the behavior analytic literature since the
1970’s (e.g., Baer, 1977; DeProspero & Cohen, 1979; Jones, Weinrott, & Vaughn, 1978).
Two viewpoints have appeared in the discussion of reliability, but experimental literature
examining visual analysis reliability has not frequently appeared. One perspective states that
visual analysis offers a reliable and conservative estimate of the data, while another argues that
improvements can occur. A total of seven studies have directly examined the reliability of visual
analysis (n=7). Five of the articles found visual analysis as an unreliably method to analyze data
(DeProspero & Cohen, 1979; Fisch, 2001; Jones et al., 1978; Matyas & Greenwood, 1990). The
57
remaining two studies suggest visual analysis can reliably analyze data (Bobrovitz &
Ottenbacher, 1998; Khang, Chung, Gutshall, Pitts, Kao, & Girolami, 2010).
Opposing Viewpoints
When considering the reliability of visual analysis, assumptions of the data should assist
in clarification of statistical testing. For example, inferential statistics relies on one datum
predicting the next. Single case data each operate independently from one another (Johnson &
Pennypacker, 2009). Thus, leaving other statistical tests that can determine effect size post
intervention (e.g. Hierarchical Linear Modeling) and descriptive statistics to determine a
numerical value for level, which provides an added support for analysts. Using visual analysis
alone has produced poor to adequate agreement (DeProspero & Cohen, 1979; Diller, Barry, &
Gelino, 2016; Fisch, 2001; Jones et al., 1978; Matayas & Greenwood, 1990; Ninci, Vannest,
Wilson, & Zhang, 2015).
Jones and colleagues asked raters to determine if a meaningful change occurred between
baseline and intervention(s) phases (1978). The researchers selected journal editors as
participants. Mean agreement across graphs came to .39. However, the term meaningful has
subjectivity and may not have adequately captured the type of change needed to make a decision.
A similar study published a year later calculated agreement among analysts at .61 (DeProsepero
& Cohen, 1979). Both results fall below a range of acceptable agreement and indicate that a
display without a clear demonstration of a functional relation fosters disagreement among raters.
Both experimenters used editors of behavior analytic journals (DeProspero & Cohen, 1979; Jones
et al, 1978).
A literature review examined a series of experiments that included the use of inferential
statistics and discovered raters may miss treatment effects using visual analysis alone (Fisch,
2001). Five manuscripts included for review found several important points to consider. First,
using visual analysis by itself creates a likelihood of treatment effects and trends in data. For
example, if a treatment effect and trend occur at the same time only one will likely receive
58
attention. Second, statistical tests exist that do not rely on parametric assumptions. If a statistical
analysis allows for independent observations, the quantification of effects aids assessment of the
impact of a given treatment. Third, statistical tests should not replace visual analysis (Fisch,
2001). While a given statistical test may prove effective in determining an effect, other problems
may arise. For example, determining sequencing effects becomes difficult when data appear
numerically and not visually. The author of the review published all the included articles (Fisch,
2001).
Another issue in regard to reliability comes in the form of Type 1 and 2 errors. A Type 1
error occurs when an analyst determines a functional relation when one does not exist. A Type 2
error takes place when a functional relation appears but goes undetected (Matyas & Greenwood,
1990). If visual analysis provides a conservative method to analyze data, one would conclude a
low rate of Type 1 error. To examine Type 1 and 2 errors, experimenters recruited 37 graduate
school students that had minimal experience outside of one single case design course at the
university level. Then, they asked participants to rate 27 graphs. Results showed the participants
had a high Type 1 error rate. The article demonstrated low reliability of visual analysis generally
and suggested a problem with agreement and increased Type 1 error (Matyas & Greenwood,
1990). Although the graphs included used an AB design, a Type 1 error in a FA could create
problems during intervention.
Interrater agreement has become a common way to determine the effects of an
independent variable on a behavior. Essentially two or more people examine a data set and offer
an analysis. A meta-analysis found interrater agreement (IRA) across single case designs at .76
(Ninci et al., 2015). The researchers determined that varying designs produce higher agreement
(e.g. multielement, AB). However, the researchers noted the analysis of effects may have caused
an increase in agreement. Specifically, studies that examined multielement designs included
raters who had extensive experience in single case research and the design uses visual aids which
59
has proven to assist to determine agreement. Visual supports may assist to increase IRA (Ninci et
al., 2015).
As previously identified, one article found IRA appears higher for some single case
designs (Ninci et al., 2015). However, another article examined the reliability of visual
inspection with hypothetical data presented in a multielement design (Diller et al., 2016).
Participants included 90 Board Certified Behavior Analysts (BCBAs) and 19 editors from JABA
and the Journal of Experimental Analysis of Behavior (JEAB). The experimenters manipulated
variability, trend, and mean shift of the data. Participants had to rate experimental control (Diller
et al., 2016).
Across both sets of participants the mean correlation of agreement came to .43. When
considering just editors the mean agreement across graphs rose to .69. Examining experimental
control through statistical testing BCBAs had a mean correlation of .52 and the journal editors
reported .44 (Diller et al., 2016). Ninci and colleagues (2015) hypothesized the impact of
experimental control increased agreement in their article and the present study did not find
significant results. Results from the study had lower values than other research. One possible
explanation includes the lack of environmental context (Diller et al., 2016). The results suggested
additional training in visual analysis may assist in determining effects (Diller et al., 2016). Also,
providing environmental context may have helped increase scores.
Supporting Viewpoints
Many prominent behavior analysts use visual analysis and classify it as a conservative
tool for examining behavior data (Baer, 1977; Michael, 1974; Parsonson & Baer, Kratochwill, &
Levin 1992). However, the previously mentioned analysts did not conduct experiments to test
their hypothesis. Two studies produced higher rates of agreement than previous studies
indicating that visual analysis may prove reliable (Bobrovitz & Ottenbacher, 1998; Khang,
Chung, Gutshall, Pitts, Kao, & Girolami, 2010). The importance of the findings suggest other
issues may cause low agreement and high Type 2 error. Another possibility for higher rates could
60
lie in the experience of the participants. As suggested in previous research, experience in
analyzing single case data may increase agreement (Ninci et al., 2015). For example, Khang and
colleagues had participants that held editor roles in behavior analytic journals (2010). The
authors found .89 agreement across displays.
The study replicated DeProspero and Cohen (1979) and demonstrated higher rates of
agreement across raters (Khang et al., 2010). Khang and colleagues found a .89 absolute
agreement which falls into the acceptable range (Müeller & Büttner, 1994). The participants
included journal editors with a high level of educational attainment and years of experience. The
findings suggest that visual analysis has reliability as a method of analysis. However, a key
difference between the two manuscripts occurred with the concept of experimental control (Ninci
et al., 2015). When considering experimental control, raters more clearly understood what they
needed to make valid judgments. The authors demonstrated that better understanding of the
procedures lead to enhanced decision making.
A study that investigated the difference between visual inspection and statistical tests
further supported the reliability of visual analysis (Bobrovitz & Ottenbacher, 1998). The study
used rehabilitation and health care workers analyzing data and then compared the results to
statistical analysis. Agreement occurred in 86% of cases (Bobrovitz & Ottenbacher, 1998),
supporting that visual analysis represents an adequate tool for interpretation of data. Namely,
agreement occurred at higher levels with medium to large effects. The experimenters determined
that effects can occur from both visual and statistical analysis. The manuscript tested ABAB
designs and did not examine other types of single case displays. For example, would analysts
determine similar conclusions with statistical testing with a multiple baseline design?
Additionally, variability remained low across graphs in the experiment (Bobrovitz &
Ottenbacher, 1998), supporting the view that when data remain stable, visual analysis proves
reliable.
61
Supporting Visual Analysis
Supports exist to improve visual analysis. The literature proposes use of statistical
methods, visual aids, and structured criteria for decision making (e.g., Brossart, Parker, Olson,
Mahadevan, & 2006; Nelson et al., 2016; Roane, Fisher, Kelley, Mevers, & Bouxsein, 2013).
Experimenters across studies recommend additional methods should support visual analysis and
not replace it. The use of statistical methods and visual aids has largely appeared within the
context of AB designs (Brossart et al., 2006; Parker, Cryer, & Byrns, 2006). The development of
structured criteria specifically applies to FA and multielement designs (e.g. Hagopian, Fisher,
Thompson, Owen-DeSchryver, Iwata, & Wacker, 1997).
Statistical Analysis
Statistical analysis of single case data has received attention in the literature. However,
not all analyses achieve acceptable results. For example, a statistical test that violates the
assumption of the independence of each observation would not provide effective analysis. Four
manuscripts paired statistical analysis and visual analysis (Brossart et al., 2006; Davis, Gangè,
Frederick, Alberto, Waugh, & Haardörger, 2013; Park, Marascuilo, & Gaylord-Ross, 1990;
Parker, Cryer, & Byrns, 2006).
A group of experimenters supplemented visual analysis with Hierarchical Linear
Modeling or HLM. HLM produced promising results as a support to visual analysis (Davis et al.,
2013). The statistical test provides an effective size based on the series of data. For instance,
once baseline and intervention data appear visually an effectiveness of an intervention can
establish overall treatment impact. By applying an effect size to visual display, it may help
provide added levels of analysis. Davis and colleagues demonstrated that HLM can support
visual analysis.
Whereas HLM has produced a supplement to visual analysis, other statistical tests
produced mixed results (Brossart et al., 2006). Researchers tested five different statistical
analyses and found two significant results. Each comparison relied on ten data points per phase
62
and used AB designs. The insignificant tests may have resulted from using a design that does not
conclusively demonstrate a functional relation. Statistical analysis that violates assumptions with
regard to independence of data may not provide appropriate methods to supplement single case
research designs.
Another study that tested rater decision making using AB designs found promising results
but used subjective language to determine effects (Park et al., 1990). The terms used to describe
effects included significant, nonsignificant, and unclear. The experimenters had judges determine
effects visually and then ran statistical analyses, comparing results. The judges that participated
in the study included one Ph.D. holder and four masters level behavior analysts. Each judge had
submitted one behavior analytic manuscript to a peer reviewed journal (Park et al., 1990).
Similar to Brossart and colleagues’ significant effects visually matched well with the results of
the statistical testing. As level, trend, and stability became less clear, the statistical tests and
raters did not match well (Park et al., 1990) further demonstrating that replacing visual analysis
with statistical testing may improve effectiveness.
When examining data on graphical displays, it is important to consider baseline data
before the implementation of an intervention. An increasing trend during a baseline phase may
skew results, making interpretation of data difficult if not impossible (Parker et al., 2006). Parker
et al. implemented two statistical tests to baseline trend and found significance. The Allison &
Gorman (ALLIS) and Mean and Slope Adjustment (MASAJ) methods may help with decision
making as a supplement to visual analysis. The experimenters used AB designs, which may have
influenced results because of the design type (Ninci et al., 2015).
Statistical analyses demonstrate usefulness as a support to visual analysis. When
comparing statistical tests such as HLM results to judge variability in the data, moderate
differences between baseline and intervention, and variability in the intervention may cause a
misrepresentation of results. Additionally, using HLM after an intervention may summarize the
effects of an intervention but will not help during the intervention because the statistical test
63
measures overall effects. Therefore, the emerging data from statistical tests like the application of
HLM may find their greatest utility to completed interventions rather than formative analyses.
Visual Aids
Visual aids may provide guidance to an analyst by summarizing a series of data. For
instance, a goal line would highlight where data should appear on a display when a behavior has
met a criterion. A trend line illustrates the direction of the data and the degree of change. The
behavior analytic literature has examined the effectiveness of visual aids (e.g. Normand & Bailey,
2006; Ottenbacher & Cusik, 1991). Currently, more evidence suggests that visual aids improve
accuracy in decision making (Fisher, Kelley, & Lomas, 2003; Hojem & Ottenbacher, 1988;
Nelson et al, 2016; Ottenbacher, 1993; Ottenbacher & Cusik, 1991; Van Norman, Nelson, Shin,
& Christ, 2013).
The studies that examined the effects of visual aids on decision making all indicated trend
lines assist with accuracy in interpretation. Extreme variability in a data set negatively affects
agreement and hinders effective decision making (e.g. Nelson et al., 2016). The utility of visual
aids has mostly received attention with AB and ABAB designs (Fisher, et al., 2003; Hojem &
Ottenbacher, 1988; Nelson et al., 2016; Ottenbacher, 1993; Van Norman et al., 2013). An
analysis on the reliability of visual analysis recommended using visual aids to increase agreement
(Ninci et al., 2015). Nelson and colleagues further support more training in how to omit extreme
values, which includes the use of visual aids. While AB and ABAB designs remain prominent,
consideration of other designs may produce different results (e.g. Alternating Treatments).
A study that focused on three experiments to increase agreement relating to treatment
effects found behavior therapists benefited from visual aids (Fisher et al., 2003). First, the
experimenters increased accuracy of visual inspection by training participants using the split-
middle (SM), dual-criteria (DC), and conservative dual-criteria (CDC). The use of the techniques
allowed for the placement of visual aids without having to use quantification of a data set. The
ability to detect trends increased accuracy with visual supports as a tool for decision making with
64
the DC and CDC methods (Fisher et al., 2001). Second, five staff received training in the DC
method. Third, the methods used to train the small group were replicated with a larger group.
Mean effects increased 24-44%, supporting the use of visual aids.
Hojem and Ottenbacher demonstrated an increase in accurate inspection across two
groups of participants with the use of trend lines for four of five displays (1988). Further, the
experimenters support the use of quantification to support visual inspection. The participant
group that used quantification had more extreme scores in agreement suggesting that although
some variability existed in the data sets, the numerical values augmented confidence with
responding. Although the graphical displays included used AB designs, the ability to detect
effects with added supports can make visual analysis a more reliable method of inspection.
Most that examine effectiveness of visual aids use a group of participants and test their
ability to determine effects. A literature review that examined agreement across fourteen studies
found that trend lines produced a mean value of .64 and data without trend lines .59 (Ottenbacher,
1993) indicating that across years a higher rate of agreement occurred with trend lines.
Controlling for hypothetical/real data, single case designs, and rater expertise, the study revealed
a .58 agreement across 789 raters (Ottenbacher, 1993).
Last, a group of experimenters examined the use of visual aids on continuous measures
(Van Norman et al., 2013). Participants examined graphs that had visual aids and no visual aids.
The data revealed that visual aids and trend magnitude influenced correct identification of trend.
Additionally, the degree to which progress appeared reduced correct responding. The ability to
identify relations between interventions and behavior reduce when a dramatic effect does not
appear on the display.
Structured Criteria
While visual aids assist with decision making, guidelines to make data based decision
may further support visual analysis. Structured criteria offer specific criteria to assist in accurate
interpretation of functional analysis (FA) data. Originally developed in 1997, the criteria
65
demonstrated that agreement increases across raters regardless of experience and educational
attainment (Hagopian et al.). Two studies have examined the use of structured criteria for FA
data (Hagopian et al., 1997; Roane et al., 2013). The criteria include several key features for
decision making. First, each graph has standardized construction features such as scaling (15.3
cm x 21.6 cm). Second, the control condition determines the placement of criterion lines that
assist in determination of function. Third, the number of data points in a test condition that fall
above the criterion lines determines function. Fourth, specific guidelines exist for automatically
reinforced functions (Hagopian et al., 1997).
One study tested the criteria on graduate students and found an increase in agreement
after implementation (Hagopian et al., 1997). The experimenters used 10 data points per
condition in the data, which closely mirrors traditional FA procedures (e.g. Iwata et al., 1994).
Additionally, the data appeared on equal interval line graphs that required the same construction
features to implement the criteria. The ability to provide standard construction of graphs, visual
aids (i.e. criterion lines), and a quantitative value of data points above or below a line impacted
agreement.
While ten data points per condition commonly occurs in practice, many behavior analysts
use fewer points per condition depending on the type of FA conducted or individual client need.
For instance, Trial Based Functional Analysis (TBFA) often uses less than ten data points per
condition to determine function (Ruiz & Kubina, 2017). Further, some clients do not require the
ten points per condition rule as the function of the behavior becomes apparent in five to six data
points. Thus, the structured criteria tested with FA data included less than ten data points per
condition (Roane et al., 2013).
Roane and colleagues used a similar procedures only modifying how many data points
appeared per condition. The criteria produced similar results as Hagopian et al. (.92 agreement)
suggested that guidelines in regard to decision making will increase agreement even with less
data to make a decision. Similarly, Roane and colleagues had to develop non-standard equal
66
interval graphs with the same construction features to implement the criteria. The experimental
design elements of standard graphical features further demonstrated the importance of appropriate
line graph construction.
An important element to the replication of the structured criteria to consider includes the
ability of visual aids and quantification to assist with agreement. With less information per
condition, agreement increased with added tools, supporting the use of standardized construction,
visual supports, and quantification. As a result, raters will more likely agree as to the relation
between the target behavior and what maintains it.
Both studies found increased agreement regardless of educational attainment and
experience (Hagopian et al., 1997; Roane et al., 2013). However, the structured criteria included
other benefits assisting decision making. In other words, each graph in both studies had the same
physical dimensions. For example, scaling of the axes remained constant. Another example,
included the same number of graphical displays tested in each session. Controlling for the visual
display elements may have contributed to the increased agreement as each graph provided
standardized features and rating the same number of graphs prevented potential fatigue.
Errors in Graphical Construction in Multielement Designs
The ability to analyze behavioral data relies on appropriate graphical construction. For
instance, if a graph does not scale appropriately, a trend line or structured criteria could lead to an
incorrect conclusion (e.g., moderately decreasing instead of rapidly decreasing). Three of the
seven dimensions of behavior analysis (Analytic, Effective, Generality) have a direct link to
visual displays (Kubina, Kostewicz, Brennan, & King, 2017). Further, errors in graphical
construction lessen the ability of supports for visual analysis (e.g. trend lines) and can affect
supports of some statistical tests (e.g. PND).
A review of line graph construction in behavior analytic journals demonstrated a large
amount of error (Kubina et al., 2017). The experimenters scrutinized the ratio of axes (i.e.,
proportional construction rule), data paths, tick marks, condition change lines, figure captions,
67
legibility of data points, label of each axis, and condition labels. The review examined varying
designs and did not exclusively focus on multielement designs. While previous research has
identified multielement designs as having higher rates of IRA, the construction of the graphs
studied all used a standardized construction guideline (e.g. Hagopian et al., 1997).
Previously published literature reviews on FA identified the Journal of Applied Behavior
Analysis (JABA) as the most frequent publisher of FA data (Beavers et al., 2013; Hanely et al.,
2003). A hand search of the journal from 2003-2016 produced 135 articles that included an FA
data set (See Appendix D). The year 2003 marked the initial publication of a literature review
that reported JABA as the primary publisher of FA data (Hanley et al.) while 2016 represented
the last complete calendar year where all issues of JABA were eligible for review.
Applying the construction features used by Kubina and colleagues’ each graph underwent
inspection for construction errors. If a graph included used FA methodology it met inclusion
criteria. Studies that targeted adaptive skills (e.g. exercise) did not meet inclusion criteria.
Multielement and ABAB designs met inclusion. However, not all FAs use condition change lines
thereby eliminating the feature from scoring. FA graphs then underwent inspection and each
feature had a score of correct or incorrect. Each graphical display could achieve a maximum of
10 correct construction features and a minimum of zero correct. For example, if a data set did not
include a data path that feature scored as incorrect for that display. Additionally, each feature has
a percentage of error across graphs (See Table 1).
FA graphs had a lower error than the overall literature examined by Kubina and
colleagues (See Table 1). Most common construction errors included the 2:3 proportional
construction rule of the axis and spacing of tick marks. The 2:3 rule represents the scaling of the
y-axis in comparison to the x-axis. Tick marks should have even spacing and appear on the
outside of the graph. Of the 135 manuscripts reviewed 118 included 1-2+ errors (See Table 2).
Seventeen of the manuscripts contained no error. The overall results indicate that an FA graph
will more likely contain an error in construction than not.
68
FA data sets differ from those reviewed by Kubina et al. by using multielement designs
and provide an assessment rather than experimental effects. Multielement designs use
comparisons to demonstrate variables maintaining challenging behavior. Essentially, FA data
compare test conditions against a control condition. When a test condition appears higher than a
control, analysts can determine the consequence delivered maintained a challenging behavior
(Betz & Fischer, 2011).
While error rates remain high with multielement designs the lower error rate in
comparison to the results of Kubina et al. should receive further examination. Other than the
purpose of the design in comparison to other types of single case graphs, the construction features
of each display should follow the same rules. Further examination may assist in creating a lower
error rate across designs.
Conclusion
The present literature review has three important conclusions. First, visual analysis can
provide accurate conclusions with respect to effects of an independent variable on a dependent
variable. However, when extreme values and/or variability occur in a data set then resolution of
effects becomes compromised. Further, the type of design used impacts determination of effects.
As suggested by one team of experimenters, some single case designs have increased agreement
(Ninci et al., 2015).
Second, while some believe visual analysis functions effectively, the literature largely
supports the need for improvement. Namely, using visual aids and quantification to increase
agreement. Additionally, having criteria with quantification and standardized construction will
increase agreement. The criteria have visual aids in the form of lines that help analysts determine
the location of one datum in comparison to other data points.
Third, when reviewing FA graphs in JABA error rates in construction occur frequently.
Based on the previously sourced literature it would appear as if using visual aids and criterion
lines becomes difficult if manipulations of graphical displays occur. Further, the manuscripts
69
included in peer reviewed journals serve as models for those entering the field of behavior
analysis. If those models violate construction rules future behavior analysts will most likely
continue creating incorrect displays.
A graphical display that incorporates standardized features, visual aids, and quantification
may provide more clarity without the need to train on additional procedures (e.g. statistical
formulas). The field of behavior analysis continues to grow with new analysts. The multielement
design lends itself well for assessment of challenging behavior and improving the likelihood of
correct identification of results is critical.
70
Reference *Indicates study was included for review Baer, D. M. (1977). Perhaps it would be better not to know everything. Journal of
Applied Behavior Analysis, 10(1), 167-172.
Beavers, G. A., Iwata, B. A., & Lerman, D. C. (2013). Thirty years of research on the
functional analysis of problem behavior. Journal of Applied Behavior Analysis, 46, 1-21.
doi: 10.1002/jaba.30
*Brossart, D. F., Parker, R. I., Olson, E. A., & Mahadevan, L. (2006). Relationship
between visual analysis and five statistical analyses in a simple AB single-case research
design. Behavior Modificaiton, 30, 531-563. doi: 10.11.77/0145445503261167
Cooper, J., O., Heron, E., T., Heward, W. L. (2007). Applied Behavior Analysis (2nd ed.). Upper
Saddle River, N.J.: Pearson Prentice Hall.
*Davis, D. H., Gagnè, P., Fredrick, L. D., Alberto, P. A., Waugh, R. E., & Haardörfer, R.
(2013). Augmenting visual analysis in single-case research with hierarchical linear
modeling. Behavior Modification, 37, 62-89. doi: 10.1177/0145445512453734
*DePropsero, A., & Cohen, S. (1979). Inconsistent visual analyses of intrasubject data.
Journal of Applied Behavior Analysis, 12, 573-579. doi: 10.1901/jaba.1979.12-573
*Diller, J. W., Barry, R. J., & Gelino, B., W. (2016). Visual analysis of data in a multielement
design. Journal of Applied Behavior Analysis, 49, 980-985. doi: 10.1002/jaba.325
*Hagopian, L. P., Fisher, W. W., Thompson, R. H., Owen-DeSchryver, J., Iwata, B. A.,
& Wacker, D. P. (1997). Toward the development of structured criteria for interpretation
of functional analysis data. Journal of Applied Behavior Analysis, 30, 313-326.
doi: 10.1901/jaba.1997.30-313
Hanley, G. P., Iwata, B. A., & McCord, B. E. (2003). Functional analysis of problem
behavior: A review. Journal of Applied Behavior Analysis, 36, 147-185.
doi:10.1901/jaba.2003.
71
*Jones, R. R., Weinrott, M. R., & Vaught, R. S. (1978). Effects of serial dependency on
the agreement between visual and statistical inference. Journal of Applied Behavior
Analysis, 11, 277-283.
*Kahng, S. W., Chung, K.-M., Gutshall, K., Pitts, S. C., Kao, J., & Girolami, K. (2010).
Consistent visual analyses of intrasubject data. Journal of Applied Behavior
Analysis, 43, 35–45. doi: 10.1901/jaba.2010.43-35
Kubina, R. M., Kostewicz, D. E., Brennan, K. M., & King, S. A. (2017). A Critical Review of
Line Graphs in Behavior Analytic Journals. Educational Psychology Review, 29, 583-
598. DOI 10.1007/s10648-015- 9339-x
Michael, J. (1974). Statistical inference for individual organism research: mixed
blessing or curse? 1. Journal of Applied Behavior Analysis, 7(4), 647-653.
*Nelson, P. M., Van Norman, E. R., & Christ, T. J. (2016). Visual Analysis Among
Novices: Training and Trend Lines as Graphic Aids. Contemporary School Psychology,
1-10.
*Ninci, J., Vannest, K. J., Willson, V., & Zhang, N. (2015). Interrater agreement between
visual analysts of single-case data: A meta-analysis. Behavior modification, 39(4), 510-
541.
*Ottenbacher, K. J. (1993). Interrater agreement of visual analysis in single-subject
decisions: Quantitative review and analysis. American Journal on Mental Retardation.
*Ottenbacher, K. J., & Cusick, A. (1991). An empirical investigation of interrater
agreement for single-subject data using graphs with and without trend lines. Journal of
the Association for Persons with Severe Handicaps, 16(1), 48-55.
*Parker, R. I., Cryer, J., & Byrns, G. (2006). Controlling baseline trend in single-case
research. School Psychology Quarterly, 21, 418-443.
72
*Parsonson, B. S., Baer, D. M., Kratochwill, T. R., & Levin, J. R. (1992). The visual
analysis of data, and current research into the stimuli controlling it. Single-case research
design and analysis: New directions for psychology and education, 15-40.
*Roane, H. S., Fisher, W. W., Kelley, M. E., Mevers, J. L., & Bouxsein, K. J. (2013).
Using modified visual-inspection criteria to interpret functional analysis outcomes.
Journal of Applied Behavior Analysis, 46, 130-146. doi: 10.1002/jaba.13
*Van Norman, E. R., Nelson, P. M., Shin, J. E., & Christ, T. J. (2013). An evaluation of
the effects of graphic aids in improving decision accuracy in a continuous treatment
design. Journal of Behavioral Education, 22(4), 283-301.
73
References for Question 3
Addison, L. R., Piazza, C. C., Patel, M. R., Bachmeyer, M. H., Rivas, K. M., Milnes, S.
M., & Oddo, J. (2012). A comparison of sensory integrative and behavioral therapies as treatment for pediatric feeding disorders. Journal of Applied Behavior Analysis, 45(3), 455–471. http://doi.org/10.1901/jaba.2012.45-455
Ahearn, W. H., Clark, K. M., MacDonald, R. P., & In Chung, B. (2007). Assessing and
treating vocal stereotypy in children with autism. Journal of Applied Behavior Analysis, 40(2), 263–275. http://doi.org/10.1901/jaba.2007.30-06
Allen, M. B., Baker, J. C., Nuernberger, J. E., & Vargo, K. K. (2013). Precursor manic
behavior in the assessment and treatment of episodic problem behavior for a woman with a dual diagnosis. Journal of Applied Behavior Analysis, 46(3), 685- 688.
Allison, J., Wilder, D. A., Chong, I., Lugo, A., Pike, J., & Rudy, N. (2012). A
comparison of differential reinforcement and noncontingent reinforcement to treat food selectivity in a child with autism. Journal of Applied Behavior Analysis, 45(3), 613–617. http://doi.org/10.1901/jaba.2012.45-613
Armstrong, A., Knapp, V. M., & McAdam, D. B. (2014). Functional analysis and
treatment of the diurnal bruxism of a 16-year-old girl with autism. Journal of Applied Behavior Analysis, 47(2), 415-419. DOI: 10.1002/jaba.122
Athens, E. S., & Vollmer, T. R. (2010). An investigation of differential reinforcement of
alternative behavior without extinction. Journal of Applied Behavior Analysis, 43(4), 569–589. http://doi.org/10.1901/jaba.2010.43-569
Athens, E. S., Vollmer, T. R., Sloman, K. N., & St. Peter Pipkin, C. (2008). An analysis
of vocal stereotypy and therapist fading. Journal of Applied Behavior Analysis, 41(2), 291–297. http://doi.org/10.1901/jaba.2008.41-291
74
Bachmeyer, M. H., Piazza, C. C., Fredrick, L. D., Reed, G. K., Rivas, K. D., & Kadey, H.
J. (2009). Functional analysis and treatment of multiply controlled inappropriate mealtime behavior. Journal of Applied Behavior Analysis, 42(3), 641–658. http://doi.org/10.1901/jaba.2009.42-641
Baker, J. C., Hanley, G. P., & Mark Mathews, R. (2006). Staff-Administered functional
analysis and treatment of aggression by an elder with dementia. Journal of Applied Behavior Analysis, 39(4), 469–474. http://doi.org/10.1901/jaba.2006.80- 05
Barretto, A., Wacker, D. P., Harding, J., Lee, J., & Berg, W. K. (2006). Using
telemedicine to conduct behavioral assessments. Journal of Applied Behavior Analysis, 39(3), 333–340. http://doi.org/10.1901/jaba.2006.173-04
Berg, W. K., Wacker, D. P., Cigrand, K., Merkle, S., Wade, J., Henry, K., & Wang, Y.-C.
(2007). Comparing functional analysis and paired-choice assessment results in classroom settings. Journal of Applied Behavior Analysis, 40(3), 545–552. http://doi.org/10.1901/jaba.2007.40-545
Berg, W. K., Wacker, D. P., Ringdahl, J. E., Stricker, J., Vinquist, K., Kumar Dutt, A. S.,
Dolezal, D., Luke, J., Kemmerer, L., & Mews, J. (2016). An integrated model for guiding the selection of treatment components for problem behavior maintained by automatic reinforcement. Journal of Applied Behavior Anlaysis, 49(3), 617- 638. DOI: 10.1002/jaba.303
Betz, A. M., Fisher, W. W., Roane, H. S., Mintz, J. C., & Owen, T. M. (2013). A
component analysis of schedule thinning during functional communication training. Journal of Applied Behavior Analysis, 46(1), 219-241.
Bloom, S. E., Iwata, B. A., Fritz, J. N., Roscoe, E. M., & Carreau, A. B. (2011).
Classroom application of trial-based functional analysis. Journal of Applied Behavior Analysis, 44(1), 19–31. http://doi.org/10.1901/jaba.2011.44-19
75
Borrero, C. S. ., & Borrero, J. C. (2008). Descriptive and experimental analyses of
potential precursors to problem behavior. Journal of Applied Behavior Analysis, 41(1), 83–96. http://doi.org/10.1901/jaba.2008.41-83
Borrero, C. S. ., & Vollmer, T. R. (2006). Experimental analysis and treatment of
multiply controlled problem behavior: A systematic replication and extension. Journal of Applied Behavior Analysis, 39(3), 375–379. http://doi.org/10.1901/jaba.2006.170-04
Bowman, L. G., Hardesty, S. L., & Mendres-Smith, A. E. (2013). A functional analysis
of crying. Journal of Applied Behavior Analysis, 46(1), 317-321. Call, N. A., Pabico, R. S., Findley, A. J., & Valentino, A. L. (2011). Differential
reinforcement with and without blocking as treatment for elopement. Journal of Applied Behavior Analysis, 44(4), 903–907. http://doi.org/10.1901/jaba.2011.44- 903
Call, N. A., Pabico, R. S., & Lomas, J. E. (2009). Use of latency to problem behavior to
evaluate demands for inclusion in functional analyses. Journal of Applied Behavior Analysis, 42(3), 723–728. http://doi.org/10.1901/jaba.2009.42-723
Call, N. A., Wacker, D. P., Ringdahl, J. E., & Boelter, E. W. (2005). Combined
antecedent variables as motivating operations within functional analyses. Journal of Applied Behavior Analysis, 38(3), 385–389. http://doi.org/10.1901/jaba.2005.51-04
Call, N. A., Wacker, D. P., Ringdahl, J. E., Cooper-Brown, L. J., & Boeiter, E. W.
(2004). An assessment of antecedent events influencing noncompliance in an outpatient clinic. Journal of Applied Behavior Analysis, 37(2), 145–157. http://d oi.org/10.1901/jaba.2004.37-145
76
Camp, E. M., Iwata, B. A., Hammond, J. L., & Bloom, S. E. (2009). Antecedent versus
consequent events as predictors of problem behavior. Journal of Applied Behavior Analysis, 42(2), 469–483. http://doi.org/10.1901/jaba.2009.42-469
Carter, S. L. (2010). A comparison of various forms of reinforcement with and without
extinction as treatment for escape-maintained problem behaviors. Journal of Applied Behavior Analysis, 43(3), 543–546. http://doi.org/10.1901/jaba.2010.43- 543
Colón, C. L., Ahearn, W. H., Clark, K. M., & Masalsky, J. (2012). The effects of verbal
operant training and response interruption and redirection on appropriate and inappropriate vocalizations. Journal of Applied Behavior Analysis, 45(1), 107– 120. http://doi.org/10.1901/jaba.2012.45-107
Contrucci Kuhn, S. A., & Triggs, M. (2009). Analysis of social variables when an initial
functional analysis indicates automatic reinforcement as the maintaining variable for self-injurious behavior. Journal of Applied Behavior Analysis, 42(3), 679–683. http://doi.org/10.1901/jaba.2009.42-679
DeLeon, I. G., Arnold, K. L., Rodriguez-Catter, V., & Uy, M. L. (2003).
Covariation between bizarre and nonbizarre speech as a function of the content of verbal attention. Journal of Applied Behavior Analysis, 36(1), 101–104. http://doi.org/10.1901/jaba.2003.36-101
Dolezal, D. N., & Kurtz, P. F. (2010). Evaluation of combined antecedent variables on
functional analysis results and treatment of problem behavior in a school setting. Journal of Applied Behavior Analysis, 43(2), 309–314. http://doi.org/10.1901/jaba.2010.43-309
Dorey, N. R., Rosales-Ruiz, J., Smith, R., & Lovelace, B. (2009). Functional analysis and
treatment of self-injury in a captive olive baboon. Journal of Applied Behavior Analysis, 42(4), 785–794. http://doi.org/10.1901/jaba.2009.42-785
77
Dozier, C. L., Iwata, B. A., Wilson, D. M., Thomason-Sassi, J. L., & Roscoe, E. M.
(2013). Does supplementary reinforcement of stereotypy facilitate extinction? Journal of Applied Behavior Analysis, 46(1), 242-255.
Dracobly, J. D., & Smith, R. G. (2012). Progressing from identification and functional
analysis of precursor behavior to treatment of self injurious behavior. Journal of Applied Behavior Analysis, 45(2), 361–374. http://doi.org/10.1901/jaba.2012.45- 361
Dupuis, D. L., Lerman, D. C., Tsami, L., & Shireman, M. L. (2015). Reduction of
aggression evoked by sounds using noncontingent reinforcement and time-out. Journal of Applied Behavior Analysis, 48(3), 669-674. DOI: 10.1002/jaba.220
Dwyer-Moore, K. J., & Dixon, M. R. (2007). Functional analysis and treatment of
problem behavior of elderly adults in long-term care. Journal of Applied Behavior Analysis, 40(4), 679–683. http://doi.org/10.1901/jaba.2007.679-683
Ebanks, M. E., & Fisher, W. W. (2003). Altering the timing of academic prompts to treat
destructive behavior maintained by escape. Journal of Applied Behavior Analysis, 36(3), 355–359. http://doi.org/10.1901/jaba.2003.36-355
Eluri, Z., Andrade, I., Trevino, N., & Mahmoud, E. (2016). Assessment and treatment of
problem behavior maintained by mand compliance. Journal of Applied Behavior Analysis, 49(2), 383-387. DOI: 10.1002/jaba.296
Fahmie, T. A, Iwata, B. A, Querim, A. C., & Harper, J. M. (2013). Test-specific control
conditions for functional analyses. Journal of applied behavior analysis, 46(1), 61-70. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/24114085
78
Falcomata, T. S., Roane, H. S., Hovanetz, A. N., Kettering, T. L., & Keeney, K. M.
(2004). An evaluation of response cost in the treatment of inappropriate vocalizations maintained by automatic reinforcement. Journal of Applied Behavior Analysis, 37(1), 83–87. http://doi.org/10.1901/jaba.2004.37-83
Falcomata, T. S., Wacker, D. P., Ringdahl, J. E., Vinquist, K., & Dutt, A. (2013). An
evaluation of generalization of mands during functional communication training. Journal of Applied Behavior Analysis, 46(2), 444-454.
Fisher, W. W., DeLeon, I. G., Rodriguez-Catter, V., & Keeney, K. M. (2004). Enhancing
the effects of extinction on attention-maintained behavior through noncontingent delivery of attention or stimuli identified via a competing stimulus assessment. Journal of Applied Behavior Analysis, 37(2), 171–184. http://doi.org/10.1901/jaba.2004.37-171
Fisher, W. W., Greer, B. D., Fuhrman, A. M., & Querim, A. C. (2015). Using multiple
schedules during functional communication training to promote rapid transfer of treatments effects. Journal of Applied Behavior Analysis, 48(4), 713-733. DOI: 10.1002/jaba.254
Fisher, W. W., Greer, B. D., Romani, P. W., Zangrillo, A. N., & Owen, T. M. (2016). Comparisons of synthesized and individual reinforcement contingencies during functional analysis. Journal of Applied Behavior Analysis, 49(3), 596-616. DOI: 10.1002/jaba.314 Fisher, W. W., Rodriguez, N. M., & Owen, T. M. (2013). Functional assessment and
treatment of perseverative speech about restricted topics in an adolescent with Asperger syndrome. Journal of Applied Behavior Analysis, 46(1), 307-311.
Fritz, J. N., Iwata, B. A., Hammond, J. L., & Bloom, S. E. (2013). Experimental analysis
of precursors to severe problem behavior. Journal of Applied Behavior Analysis, 46(1), 101-129.
79
Fritz, J. N., Iwata, B. A., Rolider, N. U., Camp, E. M., & Neidert, P. L. (2012). Analysis
of self-recording in self-management interventions for stereotypy. Journal of Applied Behavior Analysis, 45(1), 55–68. http://doi.org/10.1901/jaba.2012.45-55
Gabor, A. M., Fritz, J. N., Roath, C. T., Rothe, B. R., & Gourley, D. A. (2016). Caregiver
preference for reinforcement-based interventions for problem behavior maintained by positive reinforcement. Journal of Applied Behavior Analysis, 49(2), 215-227. DOI: 10.1002/jaba.286
Gardner, A. W., Wacker, D. P., & Boelter, E. W. (2009). An evaluation of the interaction
between quality of attention and negative reinforcement with children who display escape- maintained problem behavior. Journal of Applied Behavior Analysis, 42(2), 343–348. http://doi.org/10.1901/jaba.2009.42-343
Gouboth, D., Wilder, D. A., & Booher, J. (2007). The effects of signaling stimulus
presentation during noncontingent reinforcement. Journal of Applied Behavior Analysis, 40(4), 725–730. http://doi.org/10.1901/jaba.2007.725-730
Grauvogel-MacAleese, A. N., & Wallace, M. D. (2010). Use of peer mediated
intervention in children with attention deficit hyperactivity disorder. Journal of Applied Behavior Analysis, 43(3), 547–551. http://doi.org/10.1901/jaba.2010.43- 547
Greer, B. D., Neidert, P. L., Dozier, C. L., Payne, S. W., Zonneveld, K. L. M., & Harper,
A. M. (2013). Functional analysis and treatment of problem in behavior in early education classrooms. Journal of Applied Behavior Analysis, 46(1). 289-295.
Grow, L. L., Kelley, M. E., Roane, H. S., & Shillingsburg, M. A. (2008). Utility of
extinction-induced response variability for the selection of mands. Journal of Applied Behavior Analysis, 41(1), 15–24. http://doi.org/10.1901/jaba.2008.41-15
80
Hagopian, L. P., Bruzek, J. L., Bowman, L. G., & Jennett, H. K. (2007). Assessment and
treatment of problem behavior occasioned by interruption of free-operant behavior. Journal of Applied Behavior Analysis, 40(1), 89–103. http://doi.org/10.1901/jaba.2007.63-05
Hagopian, L. P., Contrucci Kuhn, S. A., Long, E. S., & Rush, K. S. (2005). Schedule
thinning following communication training: Using competing stimuli to enhance tolerance to decrements in reinforcer density. Journal of Applied Behavior Analysis, 38(2), 177–193. http://doi.org/10.1901/jaba.2005.43-04
Hagopian, L. P., Toole, L. M., Long, E. S., Bowman, L. G., & Lieving, G. A. (2004). A
comparison of dense-to-lean and fixed lean schedules of alternative reinforcement and extinction. Journal of Applied Behavior Analysis, 37(3), 323–338. http://doi.org/10.1901/jaba.2004.37-323
Hammond, J. L., Iwata, B. A., Fritz, J. N., & Dempsey, C. M. (2011). Evaluation of fixed
momentary dro schedules under signaled and unsignaled arrangements. Journal of Applied Behavior Analysis, 44(1), 69–81. http://doi.org/10.1901/jaba.2011.44-69
Hammond, J. L., Iwata, B. a, Rooker, G. W., Fritz, J. N., & Bloom, S. E. (2013). Effects
of fixed versus random condition sequencing during multielement functional analyses. Journal of applied behavior analysis, 46(1), 22-30. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/24114082
Hanley, G. P., Jin, C. S., Vanselow, N. R., & Hanratty, L. A. (2014). Producing
meaningful improvements in behavior of children with autism via synthesized analyses and treatements. Journal of Applied Behavior Analysis, 47(1), 16-36. DOI: 10.1002/jaba.106
81
Hanley, G. P., Piazza, C. C., Fisher, W. W., & Maglieri, K. A. (2005). On the
effectiveness of and preference for punishment and extinction components of function-based interventions. Journal of Applied Behavior Analysis, 38(1), 51– 65. http://doi.org/10.1901/jaba.2005.6-04
Harper, J. M., Iwata, B. A., Camp, E. M. (2013). Assessment and treatment of social
avoidance. Journal of Applied Behavior Analysis, 46(1), 147-160. Herscovitch, B., Roscoe, E. M., Libby, M. E., Bourret, J. C., & Ahearn, W. H. (2009). A
procedure for identifying precursors to problem behavior. Journal of Applied Behavior Analysis, 42(3), 697–702. http://doi.org/10.1901/jaba.2009.42-697
Ing, A. D., Roane, H. S., & Veenstra, R. A. (2011). Functional analysis and treatment of
coprophagia. Journal of Applied Behavior Analysis, 44(1), 151–155. http://doi.org/10.1901/jaba.2011.44-151
Ingvarsson, E. T., Kahng, S., & Hausman, N. L. (2008). Some effects of noncontingent
positive reinforcement on multiply controlled problem behavior and compliance in a demand context. Journal of Applied Behavior Analysis, 41(3), 435–440. http://doi.org/10.1901/jaba.2008.41-435
Kang, S., O’Reilly, M. F., Fragale, C. L., Aguilar, J. M., Rispoli, M., & Lang, R. (2011).
Evaluation of the rate of problem behavior maintained by different reinforcers across preference assessments. Journal of Applied Behavior Analysis, 44(4), 835– 846. http://doi.org/10.1901/jaba.2011.44-835
Kliebert, M. L., & Tiger, J. H. (2011). Direct and distal effects of noncontingent juice on
rumination exhibited by a child with autism. Journal of Applied Behavior Analysis, 44(4), 955–959. http://doi.org/10.1901/jaba.2011.44-955
82
Kodak, T., Grow, L., & Northup, J. (2004). Functional analysis and treatment of
elopement for a child with attention deficit hyperactivity disorder. Journal of Applied Behavior Analysis, 37(2), 229–232. http://doi.org/10.1901/jaba.2004.37- 229
Kodak, T., Lerman, D. C., Volkert, V. M., & Trosclair, N. (2007). Further examination
of factors that influence preference for positive versus negative reinforcement. Journal of Applied Behavior Analysis, 40(1), 25–44. http://doi.org/10.1901/jaba.2007.151-05
Kodak, T., Northup, J., & Kelley, M. E. (2007). An evaluation of the types of attention
that maintain problem behavior. Journal of Applied Behavior Analysis, 40(1), 167–171. http://doi.org/10.1901/jaba.2007.43-06
Kuhn, D. E., Chirighin, A. E., & Zelenka, K. (2010). Discriminated functional
communication: A procedural extension of functional communication training. Journal of Applied Behavior Analysis, 43(2), 249–264. http://doi.org/10.1901/jaba.2010.43-249
Kuhn, D. E., Hardesty, S. L., & Sweeney, N. M. (2009). Assessment and treatment of
excessive straightening and destructive behavior in an adolescent diagnosed with autism. Journal of Applied Behavior Analysis, 42(2), 355–360. http://doi.org/10.1901/jaba.2009.42-355
Lancaster, B. M., LeBlanc, L. A., Carr, J. E., Brenske, S., Peet, M. M., & Culver, S. J.
(2004). Functional analysis and treatment of the bizarre speech of dually diagnosed adults. Journal of Applied Behavior Analysis, 37(3), 395–399. http://doi.org/10.1901/jaba.2004.37-395
Lang, R., Davenport, K., Britt, C., Ninci, J., Garner, J., & Moore, M. (2013). Functional
analysis and treatment of diurnal bruxism. Journal of Applied Behavior Analysis, 46(1), 322-327.
83
Lang, R., Davis, T., O’Reilly, M., Machalicek, W., Rispoli, M., Sigafoos, J., Regester, A.
(2010). Functional analysis and treatment of elopement across two school settings. Journal of Applied Behavior Analysis, 43(1), 113–118. http://doi.org/10.1901/jaba.2010.43-113
Lang, R., O’Reilly, M., Machalicek, W., Lancioni, G., Rispoli, M., & Chan, J. M. (2008).
A preliminary comparison of functional analysis results when conducted in contrived versus natural settings. Journal of Applied Behavior Analysis, 41(3), 441–445. http://doi.org/10.1901/jaba.2008.41-441
Lang, R., O’Reilly, M., Lancioni, G., Rispoli, M., Machalicek, W., Chan, J. M.,
Franco, J. (2009). Discrepancy in functional analysis results across two settings: Implications for intervention design. Journal of Applied Behavior Analysis, 42(2), 393–397. http://doi.org/10.1901/jaba.2009.42-393
LaRue, R. H., Stewart, V., Piazza, C. C., Volkert, V. M., Patel, M. R., & Zeleny, J.
(2011). Escape as reinforcement and escape extinction in the treatment of feeding problems. Journal of Applied Behavior Analysis, 44(4), 719–735. http://doi.org/10.1901/jaba.2011.44-719
Lehardy, R. K., Lerman, D. C., Evans, L. M., O’Connor, A., & LeSage, D. L. (2013). A
simplified methodology for identifying the function of elopement. Journal of Applied Behavior Analysis, 46(1), 256-270.
Lindberg, J. S., Iwata, B. A., Roscoe, E. M., Worsdell, A. S., & Hanley, G. P.
(2003). Treatment efficacy of noncontingent reinforcement during brief and extended application. Journal of Applied Behavior Analysis, 36(1), 1–19. http://doi.org/10.1901/jaba.2003.36-1
84
Lomas, J. E., Fisher, W. W., & Kelley, M. E. (2010). The effects of variable time
delivery of food items and praise on problem behavior reinforced by escape. Journal of Applied Behavior Analysis, 43(3), 425–435. http://doi.org/10.1901/jaba.2010.43-425
Lomas Mevers, J. E., Fisher, W. W., Kelley, M. E., & Fredrick, L. D. (2014). The effects
of variable-time versus contingent reinforcement delivery on problem behavior maintained by escape. Journal of Applied Behavior Analysis, 47(2), 277-292. DOI: 10.1002/jaba.110
Love, J. J., Miguel, C. F., Fernand, J. K., & LaBrie, J. K. (2012). The effects of matched
stimulation and response interruption and redirection on vocal stereotypy. Journal of Applied Behavior Analysis, 45(3), 549–564. http://doi.org/10.1901/jaba.2012.45-549
Lyons, E. A., Rue, H. C., Luiselli, J. K., & DiGennaro, F. D. (2007). Brief functional
analysis and supplemental feeding for postmeal rumination in children with developmental disabilities. Journal of Applied Behavior Analysis, 40(4), 743– 747. http://doi.org/10.1901/jaba.2007.743-747
Marstellar, T. M., & St. Peter, C. C. (2014). Effects of fixed-time reinforcement
schedules on resurgence of problem behavior. Journal of Applied Behavior Analysis, 47(3), 455-469. DOI: 10.1002/jaba.134
Martin, A. L., Bloomsmith, M. A., Kelley, M. E., Marr, M. J., & Maple, T. L. (2011).
Functional analysis and treatment of human-directed undersirable behavior exhibited by a captive chimpanzee. Journal of Applied Behavior Analysis, 44(1), 139–143. http://doi.org/10.1901/jaba.2011.44-139
85
McComas, J. J., Thompson, A., & Johnson, L. (2003). The effects of presession
attention on problem behavior maintained by different reinforcers. Journal of Applied Behavior Analysis, 36(3), 297–307. http://doi.org/10.1901/jaba.2003.36-297
McKerchar, P. M., & Abby, L. (2012). Systematic evaluation of variables that contribute
to noncompliance: A replication and extension. Journal of Applied Behavior Analysis, 45(3), 607–611. http://doi.org/10.1901/jaba.2012.45-607
Miguel, C. F., Clark, K., Tereshko, L., & Ahearn, W. H. (2009). The effects of response
interruption and redirection and sertraline on vocal stereotypy. Journal of Applied Behavior Analysis, 42(4), 883–888. http://doi.org/10.1901/jaba.2009.42-883
Mitteer, D. R., Romani, P. W., Greer, B. D., & Fisher, W. W. (2015). Assessment and
treatment of pica and destruction of holiday decorations. Journal of Applied Behavior Analysis, 48(4), 912-917. DOI: 10.1002/jaba.255
Moore, J. W., & Edwards, R. P. (2003). An analysis of aversive stimuli in classroom
demand contexts. Journal of Applied Behavior Analysis, 36(3), 339–348. http://doi.org/10.1901/jaba.2003.36-339
Moore, J. W., Fisher, W. W., & Pennington, A. (2004). Systematic application and
removal of protective equipment in the assessment of multiple topographies of self-injury. Journal of Applied Behavior Analysis, 37(1), 73–77. http://doi.org/10.1901/jaba.2004.37-73
Morrison, H., Roscoe, E. M., & Atwell, A. (2011). An evaluation of antecedent exercise
on behavior maintained by automatic reinforcement using a three-component multiple schedule. Journal of Applied Behavior Analysis, 44(3), 523–541. http://doi.org/10.1901/jaba.2011.44-523
86
Mueller, M. M., Edwards, R. P., & Trahant, D. (2003). Translating multiple assessment
techniques into an intervention selection model for classrooms. Journal of
Applied Behavior Analysis, 36(4), 563–573. http://doi.org/10.1901/jaba.2003.36- 563
Najdowski, A. C., Wallace, M. D., Doney, J. K., & Ghezzi, P. M. (2003). Parental
assessment and treatment of food selectivity in natural settings. Journal of Applied Behavior Analysis, 36(3), 383–386. http://doi.org/10.1901/jaba.2003.36- 383
Najdowski, A. C., Wallace, M. D., Ellsworth, C. L., MacAleese, A. N., & Cleveland, J.
M. (2008). Functional analyses and treatment of precursor behavior. Journal of Applied Behavior Analysis, 41(1), 97–105. http://doi.org/10.1901/jaba.2008.41-97
Najdowski, A. C., Wallace, M. D., Penrod, B., Tarbox, J., Reagon, K., & Higbee, T. S.
(2008). Caregiver-Conducted experimental functional analyses of inappropriate mealtime behavior. Journal of Applied Behavior Analysis, 41(3), 459–465. http://doi.org/10.1901/jaba.2008.41-459
Neidert, P. L., Iwata, B. A., Dempsey, C. M., & Thomason-Sassi, J. L. (2013). Latency of
response during the functional analysis of elopement. Journal of Applied Behavior Analysis, 46(1), 312-316.
Northup, J., Kodak, T., Grow, L., Lee, J., & Coyne, A. (2004). Instructional influences on
analog functional analysis outcomes. Journal of Applied Behavior Analysis, 37(4), 509–512. http://doi.org/10.1901/jaba.2004.37-509
O’Reilly, M., Lang, R., Davis, T., Rispoli, M., Machalicek, W., Sigafoos, J., Didden, R.
(2009). A systematic examination of different parameters of presession exposure to tangible stimuli that maintain problem behavior. Journal of Applied Behavior Analysis, 42(4), 773–783. http://doi.org/10.1901/jaba.2009.42-773
87
O’Reilly, M. F., Sigafoos, J., Edrisinha, C., Lancioni, G., Cannella, H., Young Choi, H.,
& Barretto, A. (2006). A preliminary examination of the evocative effects of the establishing operation. Journal of Applied Behavior Analysis, 39(2), 239–242. http://doi.org/10.1901/jaba.2006.160-04
Pence, S. T., Roscoe, E. M., Bourret, J. C., & Ahearn, W. H. (2009). Relative
contributions of three descriptive methods: Implications for behavioral assessment. Journal of Applied Behavior Analysis, 42(2), 425–446. http://doi.org/10.1901/jaba.2009.42-425
Peter, C. C., Vollmer, T. R., Bourret, J. C., Borrero, C. S. ., Sloman, K. N., & Rapp, J. T.
(2005). On the role of attention in naturally occurring matching relations. Journal of Applied Behavior Analysis, 38(4), 429–443. http://doi.org/10.1901/jaba.2005.172-04
Peters, L. C., & Thompson, R. H. (2013). Some indirect effects of postivie practice
overcorrection. Journal of Applied Behavior Analysis, 46(2), 613-625. Peter Pipkin, C. S., Vollmer, T. R., & Sloman, K. N. (2010). Effects of treatment
integrity failures during differential reinforcement of alternative behavior: A translational model. Journal of Applied Behavior Analysis, 43(1), 47–70. http://doi.org/10.1901/jaba.2010.43-47
Peyton, R. T., Lindauer, S. E., & Richman, D. M. (2005). The effects of directive and
nondirective prompts on noncompliant vocal behavior exhibited by a child with autism. Journal of Applied Behavior Analysis, 38(2), 251–255. http://doi.org/10.1901/jaba.2005.125-04
88
Piazza, C. C., Fisher, W. W., Brown, K. A., Shore, B. A., Patel, M. R., Katz, R.
M., Blakely-Smith, A. (2003). Functional analysis of inappropriate mealtime behaviors. Journal of Applied Behavior Analysis, 36(2), 187–204. http://doi.org/10.1901/jaba.2003.36-187
Potoczak, K., Carr, J. E., & Michael, J. (2007). The effects of consequence manipulation
during functional analysis of problem behavior maintained by negative reinforcement. Journal of Applied Behavior Analysis, 40(4), 719–724. http://doi.org/10.1901/jaba.2007.719-724
Querim, A. C., Iwata, B. A., Roscoe, E. M., Schlichenmeyer, K. J., Ortega, J. V., & Hurl,
K. E. (2013). Functional analysis screening for problem behavior maintained by automatic reinforcement. Journal of Applied Behavior Analysis, 46(1), 47-60.
Rehfeldt, R. A., & Chambers, M. R. (2003). Functional analysis and treatment of
verbal perseverations displayed by an adult with autism. Journal of Applied Behavior Analysis, 36(2), 259–261. http://doi.org/10.1901/jaba.2003.36- 259
Rispoli, M., Camargo, S., Machalicek, W., Lang, R., & Sigafoos, J. (2014). Functional
communication training in the treatment of problem behavior maintained by access to rituals. Journal of Applied Behavior Analysis, 47(3), 580-593. DOI: 10.1002/jaba.130
Roane, H. S., Falcomata, T. S., & Fisher, W. W. (2007). Applying the behavioral
economics principle of unit price to dro schedule thinning. Journal of Applied Behavior Analysis, 40(3), 529–534. http://doi.org/10.1901/jaba.2007.40-529
Roantree, C. F., & Kennedy, C. H. (2006). A paradoxical effect of presession attention
on stereotypy: Antecedent attention as an establishing, Not an abolishing, operation. Journal of Applied Behavior Analysis, 39(3), 381–384. http://doi.org/10.1901/jaba.2006.97-05
89
Roantree, C. F., & Kennedy, C. H. (2012). Functional analysis of inappropriate social
interactions in students with asperger’s syndrome. Journal of Applied Behavior Analysis, 45(3), 585–591. http://doi.org/10.1901/jaba.2012.45-585
Rodriguez, N. M., Thompson, R. H., & Baynham, T. Y. (2010). Assessment of the
relative effects of attention and escape on noncompliance. Journal of Applied Behavior Analysis, 43(1), 143–147. http://doi.org/10.1901/jaba.2010.43-143
Rodriguez, N. M., Thompson, R. H., Schlichenmeyer, K., & Stocco, C. S. (2012).
Functional analysis and treatment of arranging and ordering by individuals with an autism spectrum disorder. Journal of Applied Behavior Analysis, 45(1), 1–22. http://doi.org/10.1901/jaba.2012.45-1
Rooker, G. W., Iwata, B. A., Harper, J. M., Fahmie, T. A., & Camp, E. M. (2011). False-
positive tangible outcomes of functional analyses. Journal of Applied Behavior Analysis, 44(4), 737–745. http://doi.org/10.1901/jaba.2011.44-737
Rooker, G. W., & Roscoe, E. M. (2005). Functional analysis of self-injurious behavior
and its relation to self-restraint. Journal of Applied Behavior Analysis, 38(4), 537–542. http://doi.org/10.1901/jaba.2005.12-05
Roscoe, E. M., Carreau, A., MacDonald, J., & Pence, S. T. (2008). Further evaluation of
leisure items in the attention condition of functional analyses. Journal of Applied Behavior Analysis, 41(3), 351–364. http://doi.org/10.1901/jaba.2008.41- 351
Roscoe, E. M., Iwata, B. A., & Zhou, L. (2013). Assessment and treatment of chronic
hand mouthing. Journal of Applied Behavior Analysis, 46(1), 181-198. Roscoe, E. M., Kindle, A. E., & Pence, S. T. (2010). Functional analysis and treatment of
aggression maintained by preferred conversational topics. Journal of Applied Behavior Analysis, 43(4), 723–727. http://doi.org/10.1901/jaba.2010.43-723
90
Roscoe, E. M., Rooker, G. W., Pence, S. T., & Longworth, L. J. (2009). Assessing the
utility of a demand assessment for functional analysis. Journal of Applied Behavior Analysis, 42(4), 819–825. http://doi.org/10.1901/jaba.2009.42-819
Roscoe, E. M., Schlichenmeyer, K. J., & Dube, W. V. (2015). Functional analysis of
problem behavior: A systematic approach for identifying idiosyncratic variables. Journal of Applied Behavior Analysis, 48(2), 289-314. DOI: 10.1002/jaba.201
Saini, V., Greer, B. D., & Fisher, W. W. (2015). Clarifying functional analysis results:
Assessment and treatment of automatically reinforced aggression. Journal of Applied Behavior Analysis, 48(3), 315-330. DOI: 10.1002/jaba.203
Samaha, A. L., Vollmer, T. R., Borrero, C., Sloman, K., Pipkin, C. S. P., & Bourret, J.
(2009). Analyses of response-stimulus sequences in descriptive observations. Journal of Applied Behavior Analysis, 42(2), 447–468. http://doi.org/10.1901/jaba.2009.42-447
Scheithauer, M., O’Connor, J., Toby, L. M. (2015). Assessment of self-restraint using a
functional analysis of self-injury. Journal of Applied Behavior Analysis, 48(4), 907-911. DOI: 10.1002/jaba.230
Shamlian, K. D., Fisher, W. W., Steege, M. W., Cavanaugh, B. M., Samour, K., &
Querim, A. C. (2016). Evaluation of multiple schedules with naturally occurring and therapist-arranged stimuli following functional communication training. Journal of Applied Behavior Analysis, 49(2), 251-264. DOI: 10.1002/jaba.293
Slocum, S. K., & Vollmer, T. R. (2015). A comparison of positive and negative
reinforcement for compliance to treat problem behavior maintained by escape. Journal of Applied Behavior Analysis, 48(3), 563-574. DOI: 10.1002/jaba.216
91
Smith, C. M., Smith, R. G., Dracobly, J. D., & Pace, A. P. (2012). Multiple respondent
anecdotal assessments: An analysis of interrater agreement and correspondence with analogue assessment outcomes. Journal of Applied Behavior Analysis, 45(4), 779–795. http://doi.org/10.1901/jaba.2012.45-779
Suess, A. N., Wacker, D. P., Schwartz, J. E., Lustig, N., & Detrick, J. (2016). Preliminary
evidence on the use of telehealth in an outpatient behavior clinic. Journal of
Applied Behavior Analysis, 49(3), 686-692. DOI: 10.1002/jaba.305 Thompson, R. H., & Iwata, B. A. (2007). A comparison of outcomes from descriptive
and functional analyses of problem behavior. Journal of Applied Behavior Analysis, 40(2), 333–338. http://doi.org/10.1901/jaba.2007.56-06
Thomason-Sassi, J. L., Iwata, B. a, & Fritz, J. N. (2013). Therapist and setting influences
on functional analysis outcomes. Journal of Applied Behavior Analysis, 46(1), 79- 87. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/24114087
Thomason-Sassi, J. L., Iwata, B. A., Neidert, P. L., & Roscoe, E. M. (2011). Response
latency as an index of response strength during functional analyses of problem behavior. Journal of Applied Behavior Analysis, 44(1), 51–67. http://doi.org/10.1901/jaba.2011.44-51
Toussaint, K. A., & Tiger, J. H. (2012). Reducing covert self-injurious behavior
maintained by automatic reinforcement through a variable momentary dro procedure. Journal of Applied Behavior Analysis, 45(1), 179–184. http://doi.org/10.1901/jaba.2012.45-179
Travis, R., & Sturmey, P. (2010). Functional analysis and treatment of the delusional
statements of a man with multiple disabilities: A four-year follow-up. Journal of Applied Behavior Analysis, 43(4), 745–749. http://doi.org/10.1901/jaba.2010.43- 745
92
Valdovinos, M. G., Roberts, C., & Kennedy, C. H. (2004). Analogue functional analysis
of movements associated with tardive dyskinesia. Journal of Applied Behavior Analysis, 37(3), 391–393. http://doi.org/10.1901/jaba.2004.37-391
Valdovinos, M. G., Roberts, C., & Kennedy, C. H. (2005). Functional analysis of
tardive dyskinesia: Implications for assessment and treatment. Journal of Applied Behavior Analysis, 38(2), 239–242. http://doi.org/10.1901/jaba.2005.53- 04
Volkert, V. M., Lerman, D. C., Call, N. A., & Trosclair-Lasserre, N. (2009). An
evaluation of resurgence during treatment with functional communication training. Journal of Applied Behavior Analysis, 42(1), 145–160. http://doi.org/10.1901/jaba.2009.42-145
Volkert, V. M., Lerman, D. C., & Vorndran, C. (2005). The effects of reinforcement
magnitude on functional analysis outcomes. Journal of Applied Behavior Analysis, 38(2), 147–162. http://doi.org/10.1901/jaba.2005.111-04
Vorndran, C. M., & Lerman, D. C. (2006). Establishing and maintaining treatment
effects with less intrusive consequences via a pairing procedure. Journal of Applied Behavior Analysis, 39(1), 35–48. http://doi.org/10.1901/jaba.2006.57-05
Wacker, D. P., Lee, J. F., Dalmau, Y. C. P., Kopelman, T. G., Lindgren, S. D., Kuhle, J.,
Pelzel, K., et al. (2013). Conducting functional analyses of problem behavior via telehealth. Journal of Applied Behavior Analysis, 46, 31-46.
Waller, R. D., & Higbee, T. S. (2010). The effects of fixed-time escape on inappropriate
and appropriate classroom behavior. Journal of Applied Behavior Analysis, 43(1), 149–153. http://doi.org/10.1901/jaba.2010.43-149
Wilder, D. A., Harris, C., Reagan, R., & Rasey, A. (2007). Functional analysis and
treatment of noncompliance by preschool children. Journal of Applied Behavior Analysis, 40(1), 173–177. http://doi.org/10.1901/jaba.2007.44-06
93
Wilder, D. A., Normand, M., & Atwell, J. (2005). Noncontingent reinforcement as
treatment for food refusal and associated self-injury. Journal of Applied Behavior Analysis, 38(4), 549–553. http://doi.org/10.1901/jaba.2005.132-04
Wilder, D. A., Register, M., Register, S., Bajagic, V., & Neidert, P. L. (2009). Functional
analysis and treatment of rumination using fixed- time delivery of a flavor spray. Journal of Applied Behavior Analysis, 42(4), 877–882. http://doi.org/10.1901/jaba.2009.42-877
Wilder, D. A., Zonneveld, K., Harris, C., Marcus, A., & Reagan, R. (2007). Further
analysis of antecedent interventions on preschoolers’ compliance. Journal of Applied Behavior Analysis, 40(3), 535–539. http://doi.org/10.1901/jaba.2007.40- 535
Wilson, D. M., Iwata, B. A., & Bloom, S. E. (2012). Computer-assisted measurement of
wound size associated with self-injurious behavior. Journal of Applied Behavior Analysis, 45(4), 797–808. http://doi.org/10.1901/jaba.45-797
Winborn-Kemmerer, L., Ringdahl, J. E., Wacker, D. P., & Kitsukawa, K. (2009). A
demonstration of individual preference for novel mands during functional communication training. Journal of Applied Behavior Analysis, 42(1), 185–189. http://doi.org/10.1901/jaba.2009.42-185
Woods, K. E., Luiselli, J. K., Tomassone, S. (2013). Functional analysis and intervention
for chronic rumination. Journal of Applied Behavior Analysis, 46(1), 328-332. Wunderlich, K. L., & Vollmer, T. R. (2015). Data analysis of response interruption and
redirection as a treatment for vocal stereotypy. Journal of Applied Behavior Analysis, 48(4), 749-764. DOI: 10.1002/jaba.227
94
Tables Table 1: Percentage of error per feature. *Not all FA’s use Condition Change lines.
Graphical Construction Feature Percentage of Manuscripts with Error
Vertical axis labeled with a quantitative
measure; horizontal axis with time unit
4%
Vertical axis length has 2:3 ratio of length to
the horizontal axis
61%
Tick marks point outward 2%
A minimal number of evenly spaced tick
marks
43%
Tick marks have labels 13%
Data Points clearly visible 1%
Data path clearly visible 0.7%
Condition Change lines* N/A
Condition labels 0.7%
Figure caption 5%
Table 2: Number of manuscripts with 0, 1, and 2+ errors. Number of Errors Manuscripts
0 Error 17
1 Error 59
2+ Errors 59
95
AppendixERecruitment Materials
Recruitment Email:
Hello BACB Certificate Holder, Thank you for taking the time to view this email. We are seeking BCBAs to participate in a survey that examines the effects of a standardized graphical display on decision making for Board Certified Behavior Analysts. The data displayed are from Functional Analyses published in peer-reviewed manuscripts. The findings will be part of a doctoral dissertation that examines issues in decision making and graphs. Should you choose to participate please find the following information: link to tutorial, information regarding the survey, and a link to the survey, which is completed online. Please click or copy and paste the links into your browser should you decide to participate. Please contact the principal investigator should you have any questions. Sincerely, Salvador “Sal” Ruiz Link to Tutorial: https://www.youtube.com/watch?v=XPqRYFE5n8A&feature=youtu.be Link to Survey: https://pennstate.qualtrics.com/jfe/form/SV_2nQewLlGHo9m6Z7
96
Informed Consent:
Penn State University
Special Education Department
Principal Investigator: Sal Ruiz M.A., BCBA
Faculty Advisor: Richard M. Kubina Jr., Ph.D, BCBA-D
Title of Study: Effects of the Functional Analysis Celeration Chart on Decision Making for
Board Certified Behavior Analysts
You’ve been invited to participate in a research project entitled “Effects of the Functional
Analysis Celeration Chart on Decision Making for Board Certified Behavior Analysts.” The
project is supervised by Dr. Richard M. Kubina. This consent document will explain the purpose
of this research project and will go over all of the time commitments, the procedures, risks and
benefits, and contact information. Please review the information provided and feel free to contact
the PI with any additional questions.
Purpose of the Study
The study is designed to determine if a standardized graphical display called the Functional
Analysis Celeration Chart (FACC) can assist Board Certified Behavior Analysts interpret
Functional Analysis results. Additionally, demographic information will be collected to determine
preferences and experience across participants.
Participation Eligibility
All Board Certified Behavior Analysts over 18 years old can participate in the survey. Further, all
participation will be anonymous.
Time Commitment
If you decide to participate, it will take approximately 30 minutes to view the tutorial and
complete the survey. There will be no further action on your behalf after completion of the
survey.
97
Tutorial & Questions
You will be asked to view a brief (~5 minutes) tutorial and answer survey questions. The tutorial
will be available via youtube.com and a link will be embedded in an email. The survey will
include 18 graphical displays with varying types of results. Further, ten additional questions
inquiring about experience and preferences will be included in the beginning and end of the
survey. One question will request an email address if you would like to be entered for a chance to
win a Visa Gift Card ™.
Risks & Benefits
There are no known risks to participate in this study. Benefits include entering to win a $50 Visa
Gift Card™. Participants that enter their email address will be randomly selected and contacted.
Access to Data
The survey and tutorial are anonymous. If you choose to enter to win the Visa Gift Card™ your
email address will be stored in a separate password protected folder. Your responses to questions
will not be associated with your email address. The PI and faculty advisor will be the only people
to have access to the data. All data will be kept in a password-protected file on the principal
investigator’s computer. The data may be used in conference presentations and/or manuscripts for
publication in peer-reviewed journals. Your identity will not be disclosed.
Participation
You may choose to stop participation at any time for any reason. You will not be penalized in any
way by stopping participation. You will not receive any punitive consequences if you choose to
withdraw from this study.
If you have any further questions at any point you may contact the principal investigator,
Salvador Ruiz M.A., BCBA at 201-575-7470 or [email protected].
Participation in this survey indicates your consent for use of the answers you provide.
VITA
Salvador Ruiz
EDUCATION Ph.D. in Special Education 2018 The Pennsylvania State University M.A. in Psychology 2014 The Chicago School of Professional Psychology B.A. in Sociology 2008 The William Paterson University of New Jersey PROFESSIONAL CERTIFICATIONS Approved BACB Supervisor, Behavior Analyst Certification Board Board Certified Behavior Analyst (BCBA) Certificate No. 1-14-10394 Certificate for Online Teaching, The Pennsylvania State University (World Campus) Advanced Certificate, Applied Behavior Analysis, The Pennsylvania State University PROFESSIONAL EXPERIENCE Feb 2017- Present Behavior Analyst. Chartlytics LLC. State College, PA. Aug. 2014- Present Graduate Assistant (Special Education), The Pennsylvania State University. University Park, PA. Jan. 2012- Jul. 2014 Behavior Specialist, Bergen County Special Services, Paramus, New Jersey. Oct. 2009- Dec. 2011 Paraprofessional, Bergen County Special Services, Paramus, New Jersey. PUBLICATIONS Ruiz, S. & Kubina, Jr., R.M. (2017). Impact of trial based functional analysis on challenging behavior and training: A review of the literature. Behavior Analysis Research and Practice, 17, 347-356. doi: 10.1037/bar0000079 Morano, S., Ruiz, S., Hwang, J., Wertalik, J., Moeller, J., & Mulloy, A. (2017). Meta-analysis of single-case treatment effects on self-injurious behavior for individuals with autism and intellectual disabilities. Autism and Developmental Language Impairments, 2, 1-26. doi: 10.1177/239694151668839