studentresiliencesurvey: preliminaryfindingsand recommenda ons · //2 studentresiliencesurvey...

Student Resilience Survey:Preliminary Findings andRecommenda�onsJuly 2020

// 2 STUDENT RESILIENCE SURVEY

Student Resilience Survey

E�� S��Between April and June of 2020, the Student ResilienceSurvey (SRS) was administered to a total of 15,331students in grades 6-12 from 54 independent and publicschools throughout the United States. All assessmentswere done a�er schools had moved to distance learning.

Results showed that across all schools, the percentage ofstudents in each grade repor�ng clinically significantsymptoms of depression ranged from 3.9%-6.6%. Parallelrates of clinically significant symptoms of anxiety rangedfrom 4.0%-7.5%.

Analyses of risk and protec�ve factors iden�fied threevariables most strongly associated with students’symptoms: low parent rela�onship quality, low structureof days, and high levels of distrac�on. In responses toopen-ended ques�ons, students o�en highlightedapprecia�on of support and understanding from schooladults.

The report concludes with a discussion of what we atAuthen�c Connec�ons believe educators need in orderto improve student well-being and resilience, along withspecific, ac�onable recommenda�ons based on what wehave learned thus far in the pandemic. We will con�nueto carefully track students’ well-being in the fall, withupdated measures that can be implemented acrossdistance-learning and in-person formats.


TABLE OF CONTENTS

INTRODUCTION 4CHALLENGES TO STUDENT WELL-BEING AND RESILIENCE 5

UNIQUE CHALLENGES RESULTING FROM COVID-19 5

THEORETICAL FRAMEWORK FOR STUDENT RESILIENCE 6

RESILIENCE AND RELATIONSHIPS 6

MENTAL HEALTH AND WELL-BEING 8

THE STUDENT RESILIENCE SURVEY 9

SURVEY DESIGN AND VALIDATION 9

SYMPTOMS 10

RISK AND PROTECTIVE FACTORS 10QUALITATIVE ITEMS 13RELIABILITY AND VALIDITY 15ASSESSMENT AND REPORTING 16

THE 2019-2020 STUDENT RESILIENCE SURVEY 17

PARTICIPANTS AND METHODOLOGY 17

RESULTS AND FINDINGS 18

WHAT EDUCATORS NEED 25

RECOMMENDATIONS 26

ONGOING INNOVATION 27

CONCLUSION 28

REFERENCES 29


INTRODUCTION

Authen�c Connec�ons (AC)is a team of leading scien�sts,clinicians, and consultantscommi�ed to helping schoolsmeasure, track, and improvestudent well-being and resilience.

AC works with schools to improvestudent outcomes by providingvaluable tools rooted in cu�ng-edge science. This includesmeasurement based on validatedsurveys, interac�ve presenta�on ofresults, and provision of ac�onableschool-specific recommenda�onsfor prac�ces and policies.

This document highlights thefindings and results from theSpring 2020 administra�on of theStudent Resilience Survey (SRS).

It begins by reviewing the currentcontext surrounding adolescent well-being and resilience. Next, it providesan overview of the scien�fic researchunderlying the SRS, and of the designand valida�on of the survey. It thenpresents findings from the mostrecent administra�on of the SRS inSpring 2020.

The report concludes with adiscussion of what we at AC believeeducators need in order to improvestudent well-being and resilience.

MISSION STATEMENT

At Authen�c Connec�ons,we aspire to maximizewell-being and resilience

in school communi�es throughdata-driven insights.


CHALLENGES TO STUDENT WELL-BEING AND RESILIENCE

Persistent trends of increasingdepression and anxiety rates amongadolescents (ADAA, 2020; CDC, 2020;NIMH, 2017; NIMH, 2019) highlightthe need for effec�ve evidence-basedtools to measure and intervene instudent well-being and resilience.

Analyses of survey data compiledby the Na�onal Ins�tute of MentalHealth show that among adolescentsin the United States, 31.9% hadanxiety and 13.3% had depression(NIMH, 2017; 2019).

For both anxiety and depression,prevalence was higher amongfemale adolescents than males(NIMH, 2017; 2019).

Over the past few decades, ourresearch group has accumulatedevidence showing that adolescentsa�ending high-achieving schools–where pressures to achieve and excelare intense and constant– have higherrates of anxiety, depression, andsubstance use than their peers in thegeneral popula�on (see Luthar, Kumar,& Zillmer, 2019; Luthar & Latendresse,2005). This phenomenon of increasedstudent vulnerability has been shown toextend across public and privateschools, ci�es and suburbs, anddifferent regions of the country(NASEM, 2019).

At AC, we believe that there is clearly aneed to focus on improving studentwell-being and resilience generally andspecifically in high-achieving schools.

UNIQUE CHALLENGES RESULTING FROM COVID-19

The COVID-19 pandemic has resultedin unprecedented disrup�ons to thedaily lives and rou�nes of studentsand their families.

The addi�on of a public health crisisand an economic recession to thestressors adolescents already faced athome and school has considerableimplica�ons for their academic

development and mental health(Golberstein, Wen, & Miller, 2020).

AC is commi�ed to developing andvalida�ng research-based tools tomeasure the impact of prolongedschool closure and future uncertainty onstudents’ mental health and well-being,and to collabora�ng with educators todeliver effec�ve school-specificinterven�ons.

THEORETICAL FRAMEWORK FOR STUDENT RESILIENCE

RESILIENCE AND RELATIONSHIPS

The Student Resilience Survey (SRS)stems from decades of researchconducted by developmental andclinical psychologists.

In 1988, Dr. Suniya S. Luthar(Professor Emerita at ColumbiaUniversity Teachers College;Co-Founder & Chief Research Officerat AC) first published a paper onresilience with Dr. Edward F. Zigler(Sterling Professor Emeritus ofPsychology at Yale University).

Since then, Dr. Luthar has maintainedan ac�ve, produc�ve program ofresearch, with pioneering contribu�onsin the field of resilience among childrenand families (see Luthar, Cicche�, &Becker, 2000; Luthar, Crossman, &Small, 2015; NASEM, 2019).

AC applies modern techniques fordata science and analy�cs to a stronglegacy of scholarly excellencefeaturing decades of peer-reviewedscien�fic research.

Resilience is the process of adap�ngwell in the face of adversity or stress.Resilience is affected by many factorsstemming from rela�onships at home,

rela�onships at school, and individuala�ributes of students themselves(Luthar et al., 2015).


Figure 1. Components of Resilience


Our research has led to theiden�fica�on of many risk factorsand protec�ve factors that impactresilience in students across allsocioeconomic backgrounds.

Risk factors are characteris�cs ofindividual students and also of theirrela�onships with others that arenega�vely related to resilience andwell-being; examples include conflictsat home or difficul�es with peers.

Protec�ve factors are aspects ofindividual students and of theirrela�onships with others that areposi�vely associated with resilienceand well-being; examples includefeeling supported by at least oneimportant adult and having posi�veviews of the school climate.

Evidence accumulated by our researchgroup has shown that students whoappear at surface level to be resilient–for instance, those who manage tosucceed academically and sociallydespite having stressful home lives–may be hiding serious symptoms ofdepression, anxiety, and substance use(see Luthar, Doernberger, & Zigler,1993; Luthar, Kumar, & Zillmer, 2019,2020).

At AC, we know that resilience rests,fundamentally, on rela�onships;we strive to measure, track, andimprove the quality of therela�onships that ma�er most tostudents.

Table 1 presents an illustra�ve list ofcommon risk and protec�ve factorsinfluencing student well-being.

Table 1. Risk and Protec�ve Factors Influencing Student Well-Being

MENTAL HEALTH ANDWELL-BEING

At AC, we partner with educatorsto improve student mental healthand well-being by assessing rates ofclinically significant depression,anxiety, and substance use instudents, and through iden�fyingrisk and protec�ve factors that aremost associated with students’ well-being within their own communi�es.

Well-being, or mental health,refers to the absence of serioussymptoms of depression, anxiety, andsubstance use.

Evidence from our research programshows that, like resilience, well-beingis influenced by many factorsstemming from rela�onships athome, rela�onships at school, andindividual a�ributes of studentsthemselves.



THE STUDENT RESILIENCESURVEY

SURVEY DESIGN ANDVALIDATION

The Student Resilience Survey (SRS)was designed to help schools assessthe impact of disrup�ons resul�ngfrom the COVID-19 pandemic onstudent well-being and mental health.

The SRS was designed to be a shortsurvey that could be completedonline by students in approximately10 minutes.

From early April through the end ofJune 2020, the SRS was administeredto over 15,000 students at 54independent (private) and publicschools across the U.S.

The SRS is a mixed-methods surveythat includes both quan�ta�ve andopen-ended ques�ons (Luthar,Ebbert, & Kumar, in press).

Quan�ta�ve items used 5-point Likertscales to measure symptoms, riskfactors, and protec�ve factors.

Qualita�ve free-response promptswere designed to capture studentconcerns and sugges�ons regardingthe (unprecedented) changes to theirdaily lives at school and at home.

SYMPTOMS

RISK AND PROTECTIVE FACTORS

The SRSmeasured two componentsof mental health: Depression andAnxiety. For each component,five Likert-scale measures askedstudents to report how frequentlythey experienced the symptom inques�on on a 5-point scale(0 = never, 4 = very o�en).

The items were taken from theWell-BeingIndex (WBI), a psychometrically-validatedmeasure consis�ng of 25 items with fivesubscales (Luthar, Ebbert, & Kumar, 2020).Each school is able to see rates ofstudents who report clinically significantlevels of depression or anxiety, rela�ve torates in na�onal norms.

The SRS assessed two essen�alcomponents of student life duringCOVID-19: Academics andRela�onships. Prior research hasshown that both components arecri�cal for resilience (Luthar,Crossman, & Small, 2015; NASEM,2019), and both have beenthreatened by school closures.

Three Likert-scale measures werecreated to assess academics-relatedfactors specific to the COVID-19context. Learning Ability assessedhow well students felt they were ableto learn at home. Learning Focusasked students how well they wereable to focus during their onlineclasses. Time for Fun asked studentsto indicate the degree to which theirtypical days at home had specific�mes set aside for ac�vi�es thatwere fun or relaxing.

Three Likert-scale measures weredesigned to assess students’rela�onships with peers and adults fromschool during the COVID-19 schoolclosures.

Sharing with Friends assessed students’sa�sfac�on with the frequency withwhich they shared personal concernswith friends.

Sharing with Adults asked students howsa�sfied they were with the frequencywith which they shared personalconcerns with adults from school.

Concerns Heard asked students aboutthe degree to which they felt teachersand administrators were listening totheir concerns about school and doingsomething about those concerns.



Two Likert-scale measures assessedboth posi�ve and nega�ve aspects ofstudents’ rela�onships with parentsor adults at home.

Parent Support assessed the degreeto which students felt their parentsunderstood and helped manage theirfeelings.

Parent Stress asked students toreport the degree to which they felttheir parents were a source of stressfor them (Luthar, Ebbert, & Kumar, inpress).

Table 2 lists measures and samplesurvey items for each component ofstudent life.


Table 2. Measures and Sample Items by Student Life Component


QUALITATIVE ITEMS

The SRS included three open-ended free response ques�ons designed to capturestudents’ feelings and insights about issues concerning them.

Using data collected during a pilot study of the free response ques�ons, a codingtaxonomy was developed in order to capture dis�nct themes and non-overlappingcategories. The coding taxonomy was refined and validated by our team inconsulta�on with two external reviewers.

Table 4. Coding Taxonomy for Free Response Ques�ons

Table 3. Free Response Prompts

Table 4 on the preceding page presents the three overarching themes on thecoding taxonomy. Table 5 lists sub-categories within each theme.


Table 5. Coding Taxonomy Sub-Categories by Theme


Inter-rater Reliability

Two team members coded all open-ended responses, and Cohen’s (1960)kappa coefficients were calculated todetermine levels of agreement.Kappa coefficients for the freeresponse ques�ons were in thesubstan�al agreement range of0.61–0.80 (Viera & Gare�, 2005).

Content Validity

To ensure the content validity of thesurvey items and coding taxonomy,two external reviewers wereconsulted. One had significantexper�se in developing systems forcoding qualita�ve data, and the otherhad classroom teaching cer�fica�onand experience.

RELIABILITY AND VALIDITY

ASSESSMENT AND REPORTING

Figure 2. Sample interac�ve dashboard.

Interac�ve dashboards were createdto present quan�ta�ve and qualita�vefindings to each par�cipa�ng school.The interac�ve dashboards allowedschool leaders to view salient findingsfor the school overall and alsoseparately for gender, ethnicity, andgrade level subgroups.

The interac�ve dashboards alsoallowed school administrators tocompare findings for their school tona�onal norms based on AC’s datafrom the 15,331 students across theU.S. who have completed the SRS todate¹.

¹ As of July 15, 2020.



THE 2019–2020 STUDENT RESILIENCE SURVEY

The analyses presented in thisdocument are based on a sampleof 15,331 students at 54 schoolsacross the U.S. who completedthe SRS between April and June ofthe 2019–2020 academic year.

The SRS was administered virtuallyduring regular school hours byschool officials following the moveto distance learning. School leadersobtained consent from students andtheir parents/guardians, giving themthe op�on to decline to par�cipateand assuring them of dataconfiden�ality and anonymity.

The findings and recommenda�onsreported in this document areorganized by student grade level.Addi�onal detailed analyses bygender and ethnicity subgroups willbe reported in a forthcomingacademic paper (Luthar et al., inprepara�on).

PARTICIPANTS AND METHODOLOGYAcross all schools in the sample,44.6% of students were male (n=6,830),51.0% were female (n=7,817), and2.7% iden�fied as non-binary (n=414).

Of the sample of students, 62.0%iden�fied as Caucasian/White(n=9,503), 14.3% as Asian/AsianAmerican/Pacific Islander (n=2,192),7.8% as Biracial/Mul�racial (n=1,197),7.9% as African American/Black (n=1,211),5.3% as La�nx/Hispanic (n=811),1.6% as Middle Eastern (n=238), and0.4% as American Indian/Na�ve American(n=68).

Table 6. Par�cipants by Grade Level

RESULTS AND FINDINGS

SYMPTOMS

Analyses of the 2019–2020 SRS datashowed that across all schools in thesample, non-zero percentages ofstudents reported experiencingclinically significant symptoms ofDepression and Anxiety– i.e., at levelsthat warrant clinical a�en�on.

In general, rates of clinically significantsymptoms of Depression increased toa peak of 6.6% among 10th graders inthe samples before decreasing; ratesof clinically significant symptoms ofAnxiety peaked at 7.5% in 11th gradebefore decreasing.


Figure 3. Percentage of Students Repor�ng Clinically Significant Symptoms by Grade Level


RISK AND PROTECTIVE FACTORS

Analyses of the risk and protec�vefactors most strongly correlated withsymptoms iden�fied three keyvariables that significantly predictedDepression and Anxiety: Low ParentRela�onship Quality, Low Structureof Days, and High Distrac�on.

Low Parent Rela�onship Quality wassignificantly predic�ve of bothDepression and Anxiety.

Figure 4 below shows the percentageof students repor�ng low parentrela�onship quality.

The measure combines students’reported feelings of high stress and oflow support in their rela�onships withparents/guardians at home.

Figure 4. Percentage of Students Repor�ng Low Parent Rela�onship Quality by Grade Level

Low Structure of Days was significantly predic�ve of both Depression andAnxiety. Figure 5 below shows the percentage of students repor�ng lack ofstructure in their daily rou�nes.


Figure 5. Percentage of Students Repor�ng Lack of Daily Structure by Grade Level


High Distrac�on was significantly predic�ve of both Depression and Anxiety.Figure 6 below shows the percentage of students repor�ng high levels ofdistrac�on in their daily rou�nes.

Figure 6. Percentage of Students Repor�ng High Levels of Distrac�on by Grade Level


Figure 7. Percentage of Students Repor�ng Low Learning Efficacy by Grade Level

Low Learning Efficacy combines students’ reported feelings of lacking necessaryresources to complete school work at home, being unable to learn new schoolmaterials at home, and being distracted during online courses. Figure 7 belowshows the percentage of students repor�ng low learning efficacy.

QUALITATIVE RESPONSES

Figure 8 below shows response themes and topics from students’ responses tothe open-ended ques�ons about what could be improved and what is going well.Larger text indicates that the topic was men�oned more frequently by students.


Figure 8. Themes and Topics in Students’ Responses to Open-Ended Ques�ons


Table 7 presents selected student responses to the free response ques�ons.

Table 7. Excerpted Student Responses to Free Response Ques�ons


WHAT EDUCATORS NEED

At AC, we believe that in addi�on torigorous data and analyses, whatschools need most from scien�sts areac�onable school-specificrecommenda�ons that clearlyhighlight next steps.

Through our interac�ve dashboards ,we present each school with theirunique profile of strengths andweaknesses derived fromquan�ta�ve and qualita�ve studentdata on symptoms, risk factors, andprotec�ve factors.

We also iden�fy the top areas offocus– the variables shown to be moststrongly associated with studentsymptoms– to be priori�zed by eachschool.

Finally, based on our review of eachschool’s quan�ta�ve and qualita�vedata and also our decades of researchon resilience and well-being, wepresent ac�onable school-specificrecommenda�ons regarding prac�cesand policies that can address studentconcerns, needs, and sugges�ons.

RECOMMENDATIONS

While all schools are unique,some general findings andrecommenda�ons have emergedfrom our research with studentsand schools.

Four broad features were commonamong those schools that did bestin terms of mental health(demonstra�ng resilience in theface of the pandemic):

1. They fostered a strong sense ofcommunity

2. They prac�ced clear and consistentcommunica�on

3. They priori�zed mental health4. They frequently sought and

addressed feedback

The table below lists examples ofac�onable next steps schools mighttake in response to student concerns.



ONGOING INNOVATION

At AC, we are commi�ed to crea�nghigh-quality tools to help schoolsmeasure, track, and improve the well-being and resilience of all members ofthe learning community.

In addi�on to the SRS, we havedeveloped the Faculty ResilienceSurvey (FRS) to assess well-being andresilience among teachers and schoolstaff, as we believe that suppor�ngstudents also requires suppor�ng theadults they rely on.

We are con�nuing to refine theWell-Being Index (WBI) and the HighAchieving Schools Survey (HASS) andregularly disseminate new findings inpeer-reviewed journals (see Luthar,Ebbert, & Kumar, in press; Luthar,Kumar, & Zillmer, 2019; Luthar, Suh,Ebbert, & Kumar, 2020) and alsodirectly to educators (see Luthar &Kumar, 2020a; Luthar & Kumar,2020b).

CONCLUSION

Overall, across schools, there were some dis�nct features of “resilient schools”–those doing well in the face of all the disrup�ons from the pandemic and distancelearning. These schools had each fostered a suppor�ve, warm, and understandingcommunity. They were commi�ed to clarity and transparency in decision making,and to proac�vely fostering and monitoring well-being and mental health. Lookingahead to the summer and fall, they are priori�zing next steps derived from thedata, focusing on issues and subgroups most needing a�en�on within their ownschools. As circumstances con�nue to evolve, we at AC are commi�ed to helpingeducators measure and track changes to student well-being and resilience over�me.



REFERENCES

Anxiety and Depression Associa�on of America (2020). Facts and sta�s�cs. https://adaa.org/about-adaa/press-room/facts-statistics

Centers for Disease Control and Preven�on (2020, June 15). Data and sta�s�cs onchildren’s mental health. https://www.cdc.gov/childrensmentalhealth/data.html

Cohen, J. A. (1960). Coefficient of agreement for nominal scales. Educa�onal andPsychological Measurement, 20, 37-46.

Golberstein, E., Wen, H., & Miller, B. F. (2020). Coronavirus disease 2019 (COVID-19) and mental health for children and adolescents. JAMA Pediatrics. Publishedonline April 14, 2020. doi:10.1001/jamapediatrics.2020.1456

Luthar, S. S. (1999). Developmental clinical psychology and psychiatry series, Vol.41. Poverty and children's adjustment. Sage Publica�ons, Inc.

Luthar, S. S. (2006). Resilience in development: A synthesis of research across fivedecades. In D. Cicche� & D. J. Cohen (Eds.), Developmental psychopathology:Risk, disorder, and adapta�on (p. 739–795). John Wiley & Sons, Inc.

Luthar, S. S., Cicche�, D., & Becker, B. (2000). The construct of resilience: A cri�calevalua�on and guidelines for future work. Child Development, 71, 543–562.

Luthar, S. S., Crossman, E. J., & Small, P. J. (2015). Resilience and adversity. In M. E.Lamb & R. M. Lerner (Eds.), Handbook of child psychology and developmentalscience: Socioemo�onal processes (p. 247–286). John Wiley & Sons Inc.

Luthar, S. S., Doernberger, C. H., & Zigler, E. (1993). Resilience is not aunidimensional construct: Insights from a prospec�ve study of inner-cityadolescents. Development and Psychopathology, 5(4), 703-717.

Luthar, S.S., Ebbert, A.M., & Kumar, N.L. (2020). The Well-Being Index (WBI) forschools: A brief measure of adolescents’ mental health. Psychological Assessment.Advance online publica�on. http://dx.doi.org/10.1037/pas0000913

Luthar, S. S. & Kumar, N. L. (2020a). Mental health ma�ers: Fostering resilienceduring COVID-19 school closures. Na�onal Associa�on of Independent Schools,April 27 (invited webinar).


Luthar, S. S. & Kumar, N. L. (2020b). Suppor�ng educators as they supportchildren: “Why” and “how-to” from resilience research. Southern Associa�on ofIndependent Schools, July 2 (invited keynote webinar).

Luthar, S. S., Kumar, N. L., & Zillmer, N. (2019). High Achieving Schools connotesignificant risks for adolescents: Problems documented, processes implicated, anddirec�ons for interven�ons. American Psychologist. Advance online publica�on.https://doi.org/10.1037/amp0000556

Luthar, S. S., Kumar, N. L., & Zillmer, N. (2020). Teachers’ responsibili�es forstudents’ mental health: Challenges in high achieving schools. Interna�onalJournal of School & Educa�onal Psychology, 8, 119–130. https://doi.org/10.1080/21683603.2019.1694112

Luthar, S. S., & Latendresse, S. J. (2005). Children of the affluent: Challenges towell-being. Current Direc�ons inPsychological Science, 14(1), 49-53.

Luthar, S. S., Suh, B., Ebbert, A.E., & Kumar, N.L. (2020). Vulnerabili�es amongstudents in high-achieving schools: Poten�al ill-effects of pressures to be'standouts'. Adversity and Resilience Science, 1, 135–147. doi.org/10.1007/s42844-020-00009-3

Na�onal Academies of Science, Engineering, and Medicine (2019). Vibrant andHealthy Kids: Aligning Science, Prac�ce, and Policy to Advance Health Equity.Washington, DC: The Na�onal Academies Press. https://doi.org/10.17226/25466

Na�onal Ins�tute of Mental Health (2017, November). Prevalence of any anxietydisorder among adolescents. https://www.nimh.nih.gov/health/statistics/any-anxiety-disorder.shtml#part_155096

Na�onal Ins�tute of Mental Health (2019, February). Prevalence of majordepressive episode among adolescents. https://www.nimh.nih.gov/health/statistics/major-depression.shtml#part_155031

Viera, A. J., & Garre�, J. M. (2005). Understanding interobserver agreement: Thekappa sta�s�c. Family Medicine, 37(5), 360-363.

studentresiliencesurvey: preliminaryfindingsand recommenda ons · //2 studentresiliencesurvey...

Documents