yaaaaaay!. what is it? in psychology we have an interest in studying human behaviour... which...
TRANSCRIPT
YAAAAAAY!
What is it?In psychology we have an interest in studying
human behaviour... which requires research!
Research methods (or scientific methods) involves the testing of the truth of a proposition (or idea) by using careful measurement and controlled observation
One of the most popular and powerful research tools is an experiment which is a formal trial to confirm or disconfirm a hypothesis
Statistics Descriptive Statistics:Summarise, organise and describe important
features of the data so it can be more easily interpreted eg) graps, tables, calculating the mean (average) etc.
Inferential Statistics:Allow the researcher to draw conclusions, based on
evidence, about the results in the study and whether they can be generalised to a wider population
We will look more at this later on
Selection and Allocation A participant is any person or group of
people used in any kind of research study, and how they are selected and allocated to groups is very important to the study
The process of selecting participants for research is called sampling. The participants actually selected for the research form the sample, which is the portion or subset of the larger population of interest
SampleIs a group of participants selected from, and
representative of the population of interest
EG:- 20 students from year 12- 10 17 year old boys
There are 2 different ways you can select a sample:
- Random- Stratified
PopulationIs an entire group of people or animals that
belong to a particular category
EG:- All Sale College students- All year 12 students- All year 12 psychology students- 15 year old boys- Under 5 feet tall girls
Random SamplingMeans that every member of the population
of interest has an equal chance of being selected for the sample to be used in the study
Some ways this can be done include:- Putting names in a hat and drawing them out- Giving each member of the population a
number and then choosing every 3rd or 5th number for the sample
Stratified SamplingIs where the sample contains exactly the same
proportions of participants as in the population. They are divided into distinct groups (strata) and then a sample from each stratum (each group) is chosen for the sample
EG:- If your sample is 100 people and you are looking at the
relationship between age and intelligence in Australia; the number of people in each age category in the sample should be exactly the same of the proportion of ages in the population
A Stratified Random Sample is where each member of a stratum has equal chance of being selected for the sample
Allocation to Groups Experimental group: the group that is exposed
to the treatment (the independent variable)
Control group: the group that is exposed to the controlling condition. It provides a standard against which to compare the performance of the experimental group to
EG:- Chocolate before class affects performance- Experimental group: Eats a block of chocolate
before class- Control Group: Doesn’t eat the chocolate
Random Allocation (to groups)Is where the participants basically have a
50/50 chance of being in the control group or experimental group. So they have equal chance if being in either
This means that any change in behaviour most likely has something to do with the independent variable
Placebo and Experimenter EffectsPlacebo effectOccurs when a participants response is influenced by their
expectations rather than by the treatment
Experimenter effectOccurs when there is a chance that the participant’s response is
due to the actions of the experimenter rather than the independent variable. It is possible for the experimenter to unintentionally sway the participant if they want to see a particular result
Experimenter bias: happens when the person measuring the dependent variable is aware of the purpose or hypothesis and may misread the data
Eliminating the Placebo Effect
Single-blind procedure Is where the participants do not know whether
they are in the control group or the experimental group
Eliminating the Experimenter Effect
Double-blind procedureIs when neither the participant or
experimenter knows which group the participant has been allocated to. In this case the person collecting the data is usually not the experimenter
Experimental DesignIs one of the most rigorous and controlled
methods used in psychology
Is used to test the cause-effect relationship between variables
VariablesAny event, state or condition that can be varied
to see what the outcome is
Independent variable (IV): the condition that is manipulated to see what the effect on the dependent variable will be
Dependent variable (DV): is affected by the IV
EG:- IV: Whether chocolate is eaten before class- DV: Behaviour/ability to work by the students
Extraneous Variables Is an variable other than the IV that might affect
the DV
EG:- Some of the children in the class had no sleep last
night because of weather disturbances- Other examples include temperament, attitudes,
mood, motivation
The experimenter should ensure that extraneous variables are eliminated from the experiment by pre empting them
Confounding Variables When an experimenter cannot be sure whether
changes in the dependent variable were caused by the independent variable or an uncontrolled variable, and the effects of the uncontrolled variable are confused with the effects of the independent variable, it is known as a confounding variable
It is basically a second, unintended independent variable
- Extraneous variables unaccounted for lead to confounding variables
Minimising effects of extraneous variablesThere are different experimental designs that
minimise the effects of extraneous variables, these are:
Repeated measures designMatched participants design Independent groups design
- Describe the differences between these 3 experimental designs, what are the pros and cons of each (table)?
HypothesisIs a testable prediction of the relationship
between two or more characteristics of events – its basically an educated guess about what the outcome of an experiment will be
EG:- Eating chocolate before class will have a
negative impact on a child’s ability to concentrate during class
Operational HypothesisIs a defined and precise prediction oh how each
variable is measured and the effect it is expected to have on behaviour. It states how the variables will be manipulated and measured, as well as the population from which the sample has been drawn
Operational definitions:Is where an experimenter defines all the terms and
subject matter being measured by describing precisely how they are going to measure it
- See page 334 of text for examples
Null Hypothesis States that there is NO relationship between
the variables In this case the experimenter is usually
expecting the hypothesis to be disproved
EG:- It is predicted that there will be no
relationship between the amount of sugar consumes before class and ability to concentrate on a task
Non-directional hypothesisIf an researcher is unsure of the direction of
the relationship between variables they will propose a non-directional hypothesis (two-tailed)
EG:- It is predicted that there is a relationship
between the amount of chocolate consumed before class and ability to concentrate on a task
Directional HypothesisIs used when researcher is confident of the
direction of the relationship (one-tailed) they will propose a directional hypothesis
EG:- It is predicted that the consumption of chocolate
(1 100 bar or more) of chocolate will decrease the students ability to concentrate on a task
Descriptive Statistics Includes graphs, tables, tables and calculations
of mean median, mode and correlation
Task: Read pages 335-343- What are correlations and how is the strength
of a correlation measured?- What are distributions and what do they tell us
about the data?- Define central tendency and its measures:
mean, median, mode, standard deviation
Central TendencyA measure of central tendency is a number
that describes a typical score around which other scores fall
It is a measure of the middle, or average of a data set
MeanIs the average, the sum of all the number
divided by the total number of numbers
EG:- 4, 5, 6, 7, 2, 1, 10 - The sum of all these is 35- Divided by 7 = 5- Mean = 5
MedianIs the number that falls in the middle of the data setTo get the median you arrange the data from lowest
to highest value, the value in the middle is the medianEG:- 78, 95, 97, 101, 110, 127, 129- 101 is the median as there are 3 values above and 3
below it - If there are even numbers, eg:- 78, 95, 97, 101, 110, 127, 129, 131- Then you would add 101 and 110 and divide by 2 to
get the average
ModeThe mode is the most frequently occurring
value
EG:- 1, 3, 4, 4, 6, 4, 7, 8, 7, 1, 4- The mode is 4
VariabilityRefers to how spread out the scores (data)
areRange: is the difference between the highest
and lowest score – calculated by subtracting the lowest value from the highest value
Standard deviation: describes how much a score differs from the mean
CorrelationCorrelation allows us to identify the
relationship between two variables – it indicates the extent to which two variables are related, but NOT that one causes the other
The strength of a correlation is measured by a correlation coefficient which is a score between
-1 and +1The closer to +1 or -1 the score is the greater
the strength of the correlation (relationship)
No CorrelationThere is no relationship between the x and y axis
Positive Correlation
As one variable
increases the other
variable increases
Negative CorrelationWhen one variable
increases the other
variable decreases
Distribution When graphed, the distribution of data often shows a pattern:
Normal distribution: most scores fall in the middle
Positively skewed distribution: there are lots of low scores and few high scores
Negatively skewed distribution: there are lots of high scores and few low scores
Inferential Statistics Allow us to draw conclusions, and to
generalise findings about samples to the broader research population
Task: Read pages 344-345- What is a conclusion?- What are generalisations? - What does statistical significance refer to and
how is it measured?
P- value P = probability, which refers to the likelihood of an
event occurring
Statistical significance (which is measured using the p-value), gives an estimate of how often results might have occurred through chance alone
The results of this significance test are stated as a probability, known as a p-value
The scale of the p-value ranges from 0-1
P-value An result that could have occurred through
chance 5 times or less out of 100 (5%) is considered significant
- P < 0.05 In some trials (mainly involving drugs) the p-
value needs to be much lower eg) p < 0.01 or less
- Look at summary of p-values on page 345 of text
Conclusions and GeneralisationsA conclusion is a decision or judgement about
what the results from the research mean eg) hypothesis supported or not, extraneous or confounding variables
A generalisation is a decision or judgement about how widely the findings of the study can be applied – especially to other members of the population – sampling techniques must be considered, as well as extraneous and confounding variables
Ethics in psychological researchDefine the following:- The role of the researcher- Participant’s rights- Confidentiality- Voluntary participation - Informed consent - Withdrawal rights- Deception- Debriefing- Professional conduct