About this Section
This section of the website contains information about research methods commonly used in the social sciences.
There is also information relating to specific modules of social science courses, including OCR and WJEC A Level Psychology and AQA GCSE Sociology.
Use the links to explore different aspects of conducting research.
Why different social scientists use different research methods:
Different Purposes
It is important to choose the right method to collect data and this depends on what is being investigated.
For example, asking questions through an interview or questionnaire is best for finding out people's opinions or what they're thinking,
whereas an observation would be better for investigating behaviour.
Different Disciplines
Some social sciences use some methods more than others, for example: psychologists are more likely to use lab experiments,
whereas sociologists are more likely to use naturalistic observations.
This is because psychologists are generally more interested in trying to establish cause and effect relationships, which requires much more control over the research environment. On the other hand, sociologists focus more on society and it is more sensible to investigate this in the real world.
[
Go to printable version of this page]
Which Statistical Test?
The test you use to analyse your data depends on the design and level of measurement.
Design
Level of Measurement
Test
Experiment, Independent Measures / Groups
Continuous - at least ordinal
Experiment, Related Measures / Groups
Continuous - at least ordinal
Experiment, Related Measures (Repeated)
Discrete - nominal
Experiment / Association, Independent Measures / Groups
Discrete - nominal
Correlation, Independent Measures / Groups
Continuous - at least ordinal
[
Go to Statistics Page]
Levels of Measurement
The levels of measurement below are in order from weakest to strongest. When trying to establish the level of measurement used, start with Nominal data and work your way through.
For example, if you are measuring continuously, then you can assume that your data is 'at least ordinal' - this is the basic assumption for the Mann-Whitney U Test, Wilcoxon's T Test and Spearman's Rho.
You can go further than this and establish whether it is ordinal, interval or ratio level by looking at the scale you are using (time in seconds, temperature in degrees centigrade, number of words found, etc.) and how these scales work.
Nominal
- Data is measured discretely in named categories.
- For example, participants' responses are measured using a descriptive term rather than a numerical score (true / false, better / worse).
- Alternatively, participants' responses are grouped together into a category (number of participants who report seeing broken glass).
Ordinal
- Data is measured continuously.
- For example, the time it takes participants to finish a wordsearch can be measured continuously in seconds.
- Scores can also be put in rank order (i.e. it would make sense to say that participant one came first, participant four came second, and so on).
Interval
- Data is measured continuously.
- For example, the ambient temperature at which participants feel their concentration is affected can be measured continuously in degrees centigrade.
- Scores can be compared using intervals (i.e. it would make sense to say that participant one was affected at a temperature 10 degrees higher than participant four, but 10 degrees is not twice as hot as 0 degrees, so it wouldn't make sense to say that participant one was able to withstand a temperature twice as high as participant four).
Ratio
- Data is measured continuously.
- For example, the number of words found could be measured continously by counting them.
- Scores can be compared in ratios (i.e. it would make sense to say that participant one found twice as many words as participant four).
Populations and Samples
Population
A population is all the people in a certain group. Eg. The population of the UK (62 million people); the population of Maghull High School Sixth Form (220 students).
Samples
As it is virtually impossible to get every member of a population to do your research, it is important to select a few people in a way that reflects what everyone else thinks or believes. There are a number of different ways to select these samples, which should generally have at least 30 people in them, or 10% of the population being investigated. If the sample represents the population properly then you can generalise your results to the population (ie. predict how everyone else in the population would respond).
Random sample
A truly random sample means that everyone in the population has an equal chance of being selected. This is actually quite hard to achieve, but the most likely way to achieve this is with the use of a computer programme.
Stratified sample
This is the name given to a particular type of quota sampling in which the population is divided into 'layers' or ‘strata' before the sample is selected, e.g. on the basis of their social class, occupation or income. It is combined with other methods.
Snowball sample
This type of sample is used if the population is difficult to contact, e.g. young offenders or drug users. As its name suggests, it involves a researcher contacting one or two people who introduce them to others - so the snowball grows.
Opportunity / Opportunistic sample
This is probably the most widely used sampling technique (especially by students) – and involves studying a sample that is easily available - that you have the opportunity of studying e.g. people in a shopping centre or students in a canteen.
Self-selecting sample
This is also a widely used sampling technique and involves asking for volunteers to take part in a study, perhaps by placing a poster in a university café or an advertisement in a magazine.
Variables
In this world there are constants and there are variables
Constants never change (like Christmas always falling on December 25th).
Variables change or can be made to change (like Easter).
Experimental Variables
Researchers are interested in the effects these changes can have on people’s behaviour.
For example, if fruit and chocolate cost the same, and the price of chocolate is increased, will this cause people to buy more fruit?
The variables the researcher is interested in are called the independent variable and the dependent variable:
Independent Variable (IV)
→
Dependent Variable (DV)
The IV is the one that is changed or manipulated by the researcher.
In the example above this would be the price of chocolate.
→
The DV is the effect of the change to the IV.
In the example above, this would be how much chocolate / fruit was bought.
Each change that the researcher makes to the IV is known as a condition. Therefore in the above example, the conditions would be same price as fruit / more expensive than fruit.
- The IV can have two or more conditions and can be naturally occuring (gender for example).
- When there is an IV manipulated by the researcher and a measurable DV, this is known as a formal experimental design / method. If the IV is naturally occurring, the design is called a quasi-experiment (quasi meaning ‘half’).
- A correlation is not an experimental method as it has two IVs. Because of this it does not measure cause and effect (ie the effect one variable has on another) – it just describes the relationship between them.
Confounding Variables
The only problem with investigating a specific variable is that other things can change as well. If all the chocolate is out of date, then people may have switched to fruit for that reason, not because it was cheaper.
It is therefore important for the researcher to try and control other variables that might affect the results, so that they can be sure of what they are measuring.
There are a number of different types of variables that may confound (or complicate) the results:
Experimenter Variables (also called experimenter effects)
This is where the presence of the researcher themselves may affect the outcome of the experiment. This can happen in 2 different ways:
- demand characteristics: this is where the participant responds to the experiment in a certain way in order to please (or upset) the researcher.
For example, if you tell your participants that you are giving them a self-esteem questionnaire in order to show that males have higher self-esteem than females, then some of the participants may deliberately or unconsciously modify their responses in order to confirm (or contradict) your hypothesis.
The way to control demand characteristics is by using a single blind design, in which the participant is not aware of the behaviour that is expected of them (i.e. they are not told whether they are in the experimental or the control group, or they are not told what behaviour is being measured or why).
- self-fulfilling prophesy: this is where the researcher subtly (and usually subconsciously) influences the behaviour of participants.
For example, psychology students in a study were told that a certain batch of laboratory rats had been specially bred to be good a finding their way through mazes. When they compared these rats to ‘ordinary rats’, they found that they were significantly better at the mazes. In fact, the ‘maze-bright’ rats were not specially bred at all, and were really no different from the ‘ordinary’ rats. The students found the result they were looking for, even though they shouldn’t have done.
This can be controlled by using a double blind design, in which even the researcher is not aware of the behaviour that is expected of the participant.
Situational Variables
This is where the environment can affect the results because it changes during the research.
For example, if the experimental group is being tested on a hot day, and the control group on a cool day, it may make a difference to the results.
It is impossible to keep all conditions identical for all participants, but it is important to do so as much as possible.
Participant Variables
This is where there are differences between the participants in the experimental and control groups.
For example, in Bandura’s study on the Transmission of Aggression, it could have been the case that the participants who witnessed the aggressive model were more aggressive to begin with.
One way round this is to have two large groups (to minimise the effect of a rogue individual) and to allocate participants to the two groups randomly.
Another technique is to use a repeated measures design, in which the experimental and control groups consist of the same group of participants, although this can also cause problems.
Other Stuff about Variables
Variables can be continuous or discrete.
For example if gender is the variable, then it is discrete because it only has 2 (or maybe more) categories; if age is the variable then it has no definite categories and can be measured on a continuum, so is continuous.
Self Reports
A self report is a method which asks participants to report their own feelings or beliefs about something. For example, it may ask them their opinions on smoking, or how much they remember about a certain event.
Self reports can take the form of interviews or questionnaires.
Questionnaires
A questionnaire is a set of carefully produced questions that are designed to investigate something in particular (eg. attitudes towards war; personalities; memory etc.). It is sometimes called a survey, and is a descriptive method, as it does not try to explain the causes of participants’ responses, it just describes them.
Interviews
An interview usually takes place face-to-face (but can be conducted online or over the phone) and consists of the researcher asking the participant about the topic under investigation.
Structured Interviews
Where a researcher has a set of standardised questions and only asks these during the interview, this is known as a structured interview.
Unstructured Interviews
Where the researcher goes into an open discussion with the participant, with no pre-defined questions, this is known as an unstructured interview.
However, the researcher will still ask questions to prompt the participant.
Types of Questions
Questions can take a number of forms (examples are in italics):
Open Questions
An open question is one that requires a detailed response with no choices given.
How do you feel about cheese?
Closed Questions
A closed question is one that gives a number of possible answers which participants choose from, and includes rated questions.
1.
I love cheese
True
False
2.
Cheese is more important than money.
Agree
Disagree
3.
Do you eat cheese?
Yes
No
Rated Questions
Rated questions offer a number of answers, each of which has a numerical value so that the participant’s responses can be turned into a score at the end.
1.
Cheese is the most important thing in my life.
agree strongly
+2
agree
+1
neither agree nor disagree
0
disagree
-1
disagree strongly
-2
THIS IS CALLED A LIKERT SCALE
2.
Where 1 is not important and 5 is very important, please rate the following statements:
Cheese must be in a nice packet
1
2
3
4
5
Cheese must be cheap
1
2
3
4
5
3.
Please tick the box that is closest to your feelings about the following:
On psychology exams you may well be asked to describe the use of rated questions, so it is worth taking the opportunity to try them all out in a questionnaire. Normally you would only choose one style of rated question.
You will probably need to use a mixture of closed and rated questions in designing your questionnaire. Open questions are useful but do take a lot longer to analyse and can not be rated in the same way as closed and rated questions.
Therefore, if you were going to investigate whether cheese is more important to males than females, your questionnaire should be designed to provide a score that makes this easy to see: the higher the participant’s overall score, the more important they believe cheese to be.
Observations
Types of Observation
There are a number of ways to conduct an observation, and your choice will depend on the topic of your research.
Direct Observation
A direct observation is the type most often used in experiments, as the participants will be watched completing the experiment, and the researcher will record what they observe. Usually the participants are aware that they are being observed (overt or open observation).
Participant Observation
A participant observation is the type most often used in naturalistic research, as the researcher is part of the group they are observing (eg observing students in the study area). Participants may not be aware that they are being observed (covert).
Overt and Covert Observations
Sometimes it is necessary to watch participants without them knowing, as their behaviour might change if they know they are being observed. This is called a covert observation.
If participants are aware they are being observed, then it is called an overt observation.
Whatever sort of observation you do, you must not invade the privacy of participants.
Observation Schedules / Sampling Behaviour
Because observations can take up a lot of time, it is important that you decide in advance what kind of behaviours you are interested in observing and recording, and over how many occasions / how long you will observe. This will ensure that you collect the right information.
It is impossible to observe everything that the participants are doing and in fact most of their behaviour might be irrelevant to the study. Therefore we only sample some of the behaviour that takes place - there are two ways to do this, which can be used independently or combined:
Time Sampling
This involves measuring target behaviour over a period of time, such as every five seconds.
Event Sampling
This involves identifying specific events (behaviours) that might happen during the observation and only recording these.
For example, if you were going to investigate the studying habits of males and females, then you using event sampling, you would need to categorise studying behaviour and record it. Categories might include working on a computer, reading a text book, writing an essay etc.
Alternatively, you could just record the number of males and females who are studying every five seconds (time sampling), or a combination of both, where the events are recorded every five seconds rather than every time they happen.
Coding Behaviour
On the exam they might ask you to ‘describe categories of behaviour and the rating or coding system’ or to ‘consider the alternative ways of sampling the behaviour’.
For the example above, the coding system may just be a tally chart of how many males and females were doing each of the categories during the observation.
However, if you were observing flirting behaviour, then it might be necessary to rate that behaviour, similarly to the way questions are rated in a self report.
Participant
Ob 1
Ob 2
Ob 3
Ob 4
Ob 5
Ob 6
(1 = least flirtatious 10 = most flirtatious)
Inter-observer / Inter-rater Reliability
Because you will be making a value judgement about people’s behaviour, it is important to try and get some control over this. You might think someone is flirting a lot when the rest of the research team thinks they aren’t flirting at all!
As a group of researchers, you would need to all observe the same participants and come to an agreement on how flirtatious they were being. This should make the results more reliable.
This is called inter-observer reliability.
Experiments
Experiments involved studying the effects of an independent variable (IV) on a dependent variable (DV).
In order to do this, researchers set up two or more different conditions of the IV.
If these conditions occur naturally (for example, gender has the conditions male and female), then the experiment is a quasi or natural experiment.
The following table outlines the different types of experiment - what they all have in common is that they have an IV and a DV.
Design
Environment
Variables
Formal Experiment
Controlled (eg. laboratory)
Manipulated by researcher - p's can be randomly allocated to the conditions of the IV
Quasi Experiment
Natural or Controlled
IV occurs naturally
Field Experiment
Natural
Naturally occurring or maipulated by researcher
Investigations with two conditions usually have an experimental group and a control group. (eg if looking at the effectiveness of paracetomol on pain, then the experimental group would have paracetomol and the control group would have none or a placebo).
In psychology exams you will need to demonstrate that you understand different types of experimental design and their strengths and weaknesses.
Independent Conditions
(also called Independent Groups / Samples / Measures / Subjects)
See also section on Variables
Where participants only take part in one condition or the other, these are known as independent conditions. Usually this means there are two groups, one activity.
Examples:
Hypothesis: Teachers are less tolerant of recreational drug use than students.
IV: Age (2 conditions – teachers /students) Participants are selected on the basis of being one or the other.
DV: tolerance of drug use
Hypothesis: The presence of rock music will significantly improve recall of music trivia.
IV: Music (2 conditions – music / no music) Participants are randomly assigned to one or the other.
DV: Amount of information recalled.
Related Conditions
A related design can be done in two ways:
1. Repeated Measures
This is where participants experience both conditions. One group, two activities.
Example:
Hypothesis: Subjects will remember more words when they are wide awake than when they are tired.
IV: Tiredness of Participants (2 conditions – wide awake / tired) – participants do one condition then the other.
DV: recall of words
2. Matched Pairs
This is where participants only do one condition or the other, but they are matched in all ways other than the variable being tested. Two groups, two activities.
Example:
Hypothesis: Alcohol impairs judgement of car speed
IV: Alcohol consumed (2 conditions – alcohol / no alcohol) – participants are matched on age, driving experience, gender etc.
DV: Judgement of speed of car
Statistical Analysis
For the purposes of AS Level we do not conduct a statistical analysis. However, if you wish to extend your learning, you can do this independently. This is also the case for the correlational research.
You will need to know how to conduct statistical analyses for A2 Level and above.
Mann-Whitney U Test (independent design)
Wilcoxon’s T Test (related design)
Spearman’s Rho Test (correlation)
Chi-Squared Test (association)
Sign Test (related design)
Standard Deviation
Formula for calculating Standard Deviation:
An Example
Most people are comfortable with working out an average. If your shopping bill goes up and down every week and someone asks you how much you spend, you know that if you were to take your last few shopping receipts and add them all together, then divide the overall total by the number of receipts, this will give you an average for your weekly shopping bill.
My Shopping in February |
Week |
Receipt Total (£) |
1 |
102 |
2 |
112 |
3 |
102 |
4 |
100 |
Total |
416 |
Mean |
104 |
Average (mean) = £416 divided by number of weeks - 416 / 4 = £104
Alternatively, you might decide to just go with a quick shortcut and quickly scan through the receipts to find out what you most often spend. This is also a sort of average - the one described in the previous paragraph is called a
mean, whereas this one is called the
mode (the most frequent amount spent).
Average (mode) = most frequent spend = £102
There is a third type of average called a
median, but I'll leave explaining that for now and just tell you that the official name for an average is a
measure of central tendency. As the name suggests, this is a way of describing your shopping by lumping it all together and coming up with a number that indicates about how much you usually spend. This would be handy if you wanted to compare how much you spend to how much other people spend, but it's not perfect.
The problem is that averages can be highly inaccurate. For instance, yesterday I spent £202 at Costco (never been there before, so it was the novelty of it I think), but I will still need to go and do a proper grocery shop at some point. Should I decide to look back at my shopping bills for March, this £202 is going to really throw the average, making it much higher than any other month, just because of one week's unusually high bill. This could mean that in the other 3 weeks I only spend £100, but come out with an average of around £150 per week, whereas my sister could consistently spend £50 more a week than me and come out with the same average.
My Shopping (March) |
Sister's Shopping (March) |
Week |
Receipt Total (£) |
Week |
Receipt Total (£) |
1 |
102 |
1 |
151 |
2 |
104 |
2 |
152 |
3 |
302 |
3 |
153 |
4 |
98 |
4 |
150 |
Total |
606 |
Total |
606 |
Mean |
151.50 |
Mean |
151.50 |
This is where a second type of descriptive statistic comes in handy...
Standard Deviation is a type of what mathematical people call a 'measure of distribution'.
Measures of distribution look at the pattern of all the data and the simplest type is the
range, which is calculated by subtracting the lowest bill from the highest. Thus, I could show that even though my sister and I both spent on average £150 per week, the range for my shopping was £204, whereas for hers it was only £3. This shows that her shopping bills are far more consistent than mine.
So far, we've looked at the mean, mode and range, all of which are a bit limited. With the mean, as seen above, this can be very unreliable if there are any anomalous scores (my £202 Costco binge). The mode and range only look at one or two scores, not all of them, so might not spot other odd things going on in the data.
Standard Deviation is the next step up in accurately describing the numbers in a form that is much easier to read than looking at a whole table full of numbers. It's easy enough in this shopping example to scan through 4 weeks' worth of bills, but if this was a whole years' worth of shopping bills, then that wouldn't be quite so practical.
All that standard deviation does is look at each of the shopping bills and figure out as a general rule how much these go against the norm (ie how much it deviates from the average).
The first step then has to be calculating the average - in this case the mean, which we know means adding all of the shopping bills together and dividing this by the number of bills we're dealing with.
My Shopping (March) |
Week |
Receipt Total (£) |
1 |
102 |
2 |
104 |
3 |
302 |
4 |
98 |
Total |
606 |
Mean |
151.50 |
Now we need to look at how much difference there is between each bill and the average. This is likely to mean that some weeks I spent less and some weeks I spent more, so for some the difference will be a negative value and for some a positive value.
Week 1 £151.50 - £102 = £49.50
Week 2 £151.50 - £104 = £47.50
Week 3 £151.50 - £302 = £-150.50
Week 4 £151.50 - £98 = £53.50
There is a basic rule in maths that states two signs that are the same make a positive. This is quite handy, because if I was going to add up all the differences between the average and my various shopping receipts, the minuses would cancel out the pluses. Also, it doesn't matter whether I spent more or less, only that I spent a different amount to the average.
So, if we multiply a negative number by itself (square it), it becomes a positive number, although it will obviously make the difference between each shopping bill and the average bigger than it actually is, but this doesn't matter, as it can be easily reversed. This means we now have all positive numbers and can add up all the values without them cancelling each other out. Later on we can 'unsquare' the total (or find the square root).
Week 1 £49.50 x £49.50 = £2,450.25
Week 2 £47.50 x £47.50 = £2,256.25
Week 3 £-150.50 x £-150.50 = £22,650.25
Week 4 £53.50 x £53.50 = £2862.25
Total = £30,219
To sum up so far, I have:
- found the average (mean) for my weekly shopping
- found out the difference between each bill and the average (average - receipt total)
- for each bill, squared this difference to make sure that it is a positive number
- added all the differences together
Now, the number I have at this point tells me the total of all the differences, so I need to divide this by the number of shopping receipts I have to get the average difference (or variance) between my shopping and the mean.
Total = £30,219
Number of receipts = 4
Variance = Total / Number of receipts
Variance = 7,554.75
Now I have variance, I just need to reverse the squaring I did earlier and what I am left with is the standard deviation.
SD = the square root of the variance
SD = the square root of 7,554.75
SD = 86.92
What does this mean?
It is a number that, taken together with the average (mean), will easily describe what my shopping spend looks like. My average weekly shop of £150 has a standard deviation of £86.92, which now shows that I am rather inconsistent in how much I spend on shopping (it really goes up and down a lot). This is a more accurate description of my shopping habits than the mean would be on its own.
One final note...
Even the standard deviation isn't much help really. After all, I don't always spend £70.92 more or less than my average. So, descriptive statistics are helpful for presenting data in a way that patterns can be easily seen, but to really get to the bottom of what is going on, we need another beast entirely: inferential statistics.
OCR G541 Psychological Investigations
On This Page:
Links to Relevant Documents and Webpages
About Module G541: Psychological Investigations
G541 is assessed through a one hour examination, with three sections (A, B and C) – 20 minutes per section, with 20 marks for each section.
The questions are based around three of the four main methods used in psychological investigations:
- Self Reports / Questionnaires
- Observations
- Experiments
- Correlations
There is really no way of predicting which three will come up, so you need to know all four methods in good detail, including how they work, the different ways they are put into operation and their strengths and weaknesses.
You will also need to know about the different sampling methods used to select participants, how to effectively present and interpret results and how investigations could be conducted differently.
The best way to learn about psychological investigations is to conduct some for yourself. This makes it easier to understand how the methods work and some of the problems with them.
The following information is aimed at guiding you through the process of writing up those all important reports after you have completed your investigations.
For more information on conducting the studies, refer to your AS Psychology Handbook.
Some Important Notes on How to Complete Research Reports
- The report should be double line spaced - if using Microsoft Word
this can be done from the Format menu. Select all of the text to be double
spaced, choose PARAGRAPH from the FORMAT menu, and change LINE SPACING to
DOUBLE. Click OK.
- All tables and graphs must be labelled properly, with titles.
- All pages should be numbered, and preferably have your name on them (or
candidate number if later they will be peer-assessed).
- Research subjects should be referred to by a number, not their name to
maintain confidentiality.
- All other information on descriptive statistics, evaluation issues etc. is in the AS Level Psychology Handbook, available from the online document
archive.
Self Reports / Questionnaires
Aim
State here what the aim of the investigation was - what you investigated.
Participants
State here the population from which your subjects were chosen, the sampling
method (how you chose them - usually opportunistic or random), and
the make up of the sample (number, males, females, age).
Procedure
Include here:
Design
- That this used a questionnaire
- How the questions were developed (usually through a pilot study or
by brainstorming with other researchers.
- Description of the questionnaire, including the style of questions,
how responses were rated, what the ratings meant.
- Whether the investigation looked at the difference between two groups
(eg males and females)
Measurement of Variables
- What the questionnaire was measuring
- How this was measured (eg: The questionnaire measured male and female
attitudes towards studying. Each question was rated using a Likert scale
of 1 to 5, with higher scores indicating a more positive attitude towards
studying. There were ten questions on the questionnaire, which meant subjects
could score a minimum of 10, and a maximum of 50)
Procedure
- What was done, including what instructions were given and whether subjects
were informed of the reason for the research.
Controls
- Anything that was done to make sure the research was valid, including
questioning all subjects at the same time, using the same instructions
etc.
Ethics
- State the ethical considerations and what was done to ensure that the
research was ethical.
Results
- A table to show the results
- Mean scores and other descriptive statistics as relevant
(eg range of scores - range = highest total questionnaire score minus
lowest score)
- A verbal summary, including unusual results and a description of what
the graph shows (the graph should be an appendix at the end)
Evaluation / Conclusion
- Depends what was found in the investigation.
- Also include any strengths or weaknesses, how these could be addressed
in future research and what the effect of any changes would be.
Appendices
- Blank questionnaire with rating scales shown under each question
- Raw data
- Standardised instructions given to subjects
- Graphs
Observations
Aim
State here what the aim of the investigation was - what you investigated.
Participants
State here the population from which your subjects were chosen, the sampling
method (how you chose them - usually opportunistic or random), and
the make up of the sample (number, males, females,age).
Procedure
Include here:
Design
- That this was an observation and what type (participant / non-participant
and whether it was direct; also whether it was covert (subjects did not
know they were being observed)
- How observation schedule and behaviour ratings were developed (usually
through a pilot study or by brainstorming with other researchers.
- Description of the observation, including how behaviour was coded,
what the ratings / coding meant.
- Whether the investigation looked at the difference between two groups
(eg males and females)
Measurement of Variables
- What the observation was measuring
- How this was measured (eg: The observation investigated male and female
studying habits, by observing their behaviour in the study area. Studying
behaviour was coded through a tally of how many males and how many females
were studying. The researchers recorded this information at 5 minute intervals.)
Procedure
- What was done, including what instructions were given and whether subjects
were informed of the reason for the research.
Controls
- Anything that was done to make sure the research was valid, including
questioning all subjects at the same time, using the same instructions
etc.
Ethics
- State the ethical considerations and what was done to ensure that the
research was ethical.
Results
- A table to show the results
- Mean scores and other descriptive statistics as relevant.
- A verbal summary, including unusual results and a description of what
the graph shows (the graph should be an appendix at the end)
Evaluation / Conclusion
- Depends what was found in the investigation.
- Also include any strengths or weaknesses, how these could be addressed
in future research and what the effect of any changes would be.
Appendices
- Observation schedule, including any coding / rating for behaviours and
tables for completion in observation.
- Raw data
- Standardised instructions given to subjects
- Graphs
Experiments
Hypotheses
Experimental Hypothesis
There is a difference between etc. (two-tailed)
OR
Condition one will be better / more likely to do / score higher or whatever
than condition two (one-tailed)
Null Hypothesis
There is no difference etc.
Participants
State here the population from which your subjects were chosen, the sampling
method (how you chose them - usually opportunistic or random), and
the make up of the sample (number, males, females, age).
Procedure
Include here:
Design
- Whether this was an independent groups, repeated measures or matched pairs design.
- Whether it was a formal or quasi experiment.
- What method was used to collect the data (observation / questionnaire
/ interview etc.)
Measurement of Variables
- State the Independent Variable and its two (or more) conditions.
- State the Dependent Variable (the behaviour that is being affected
by the IV).
Procedure
- What was done, including what instructions were given and whether subjects
were informed of the reason for the research.
- Explain how each subject's score was calculated (eg from a rated
questionnaire).
Controls
- Anything that was done to make sure the research was valid, including
questioning all subjects at the same time, using the same instructions
etc.
Ethics
- State the ethical considerations and what was done to ensure that the
research was ethical.
Results
- A table to show the results
- Mean scores and other descriptive statistics as relevant
(eg range of scores - range = highest total questionnaire score minus
lowest score)
- A verbal summary, including unusual results and a description of what
the graph shows (the graph should be an appendix at the end)
Evaluation / Conclusion
- Depends what was found in the investigation.
- Also include any strengths or weaknesses, how these could be addressed
in future research and what the effect of any changes would be.
Appendices
- Blank questionnaire with rating scales shown under each question / Observation
schedule
- Raw data
- Standardised instructions given to subjects
- Graphs
Correlations
Hypotheses
Experimental Hypothesis
There is a correlation between variable 1 and variable 2 (two-tailed)
OR
There is a positive correlation between variable 1 and variable 2 (one-tailed)
OR
There is a negative correlation between variable 1 and variable 2 (one-tailed)
Null Hypothesis
There is no relationship between variable 1 and variable 2
(Substitute variable 1 and variable 2 in the above for the real variable 1 and variable 2 .)
Participants
State here the population from which your subjects were chosen, the sampling
method (how you chose them - usually opportunistic or random), and
the make up of the sample (number, males, females, age).
Procedure
Include here:
Design
- What method was used to collect the data (observation / questionnaire
/ interview etc.)
Measurement of Variables
- State Variable 1
- State Variable 2
- Explain what these are and how they were measured (ie Sleep was measured
through a self report questionnaire rating quality of sleep compared to
the subject's usual night's sleep)
Procedure
- What was done, including what instructions were given and whether subjects
were informed of the reason for the research.
- Explain how each subject's score was calculated (eg from a rated
questionnaire)
- Explain how this was analysed.
Controls
- Anything that was done to make sure the research was valid, including
questioning all subjects at the same time, using the same instructions
etc.
Ethics
- State the ethical considerations and what was done to ensure that the
research was ethical.
Results
- A table to show the results
- Mean scores and other descriptive statistics as relevant
(eg range of scores - range = highest total questionnaire score minus
lowest score)
- A verbal summary, including unusual results and a description of what
the scattergram shows (the scattergram should be an appendix at the end)
Evaluation / Conclusion
- Depends what was found in the investigation.
- Also include any strengths or weaknesses, how these could be addressed
in future research and what the effect of any changes would be.
Appendices
- Blank questionnaire with rating scales shown under each question / Observation
Schedule
- Raw data
- Standardised instructions given to subjects
- Scattergram
Evaluations
When evaluating your investigations you should think about:
- Any potential problems with your participants and how they were selected.
- Whether the ways in which you measured behaviour, attitudes etc. were valid (measured what they were supposed to).
- If the procedure was controlled and the same for all who took part.
- Whether your chosen data collection methods were the best ones for the job.
- How possible it would be to replicate what you did.
- How useful your findings are.
- What future studies would be useful in the area you researched.
For each point you make you need to:
- Identify whether the point is about a strength or a weakness.
- Explain it in detail, outlining why it is a strength / weakness and the effect this has on the study.
- Suggest a way to improve it (if it’s a weakness) or how this makes the research useful (if it’s a strength).
- Describe what you think the effect of any improvements would be / suggest what future research could be done.
Finally, when you conclude, you should tie up your results and evaluations, as your results might be invalid once you’ve finished with your evaluation!
Research Concepts / Evaluation Issues
Why is this page green?
Because Dougie said so!
RESEARCH CONCEPTS / EVALUATION ISSUES
Qualitative and Quantitative Data
Quantitative Data
Data that can be measured numerically (eg the number of people who answer "yes" to the question "Have you ever been bullied?") (quantity)
Qualitative Data
Data that can not be measured in numbers (eg how someone defines bullying) (quality)
Reliability and Validity
In any experiment, the DV has to be operationalised; in other words, the researcher has to find a way of measuring the DV that is reliable and valid.
Reliability
A reliable measure is one that is consistent. For example, an individual’s IQ is not supposed to vary from day to day; if you use an IQ test that keeps giving different scores, then it is not reliable. There are 4 types of reliability:
- inter-rater reliability: do different researchers get the same result if they test the same participant?
- test-retest reliability: do you get the same result if you test the same participant again?
- alternate forms reliability: do two different versions of the same test give similar results?
- split-half reliability: do the scores you get from the first part of a test match up with the scores you get from the second part?
Validity
A valid measure is one that is really measuring what it claims to be measuring. For example, does an IQ test really measure intelligence? Does the questionnaire you have designed to see how depressed people are really measure their true level of depression? There are 3 types of validity:
- face validity: on the face of it, does your measure look valid? i.e. does common-sense say that it might well be measuring what you say it is?
- criterion validity: does your measure give you similar results to other measures that are supposed to be measuring the same thing?
- construct validity: is the theory underpinning your measure valid? i.e. can you demonstrate a theoretical connection between the test you are using and the behaviour it is supposed to measure.
- internal validity: is the research measuring what it's supposed to within its own setting.
- external validity: is the research measuring what it's supposed to outside of its setting (also referred to as 'ecological validity').
Ecological Validity
This term simply refers to whether the research is relevant to the real world. If it is then it has ecological validity, if it isn’t then it doesn’t!
Generalisability
As it is usually necessary to limit research to one setting (eg one country / one place of work) and the research may only consider a small sample of people, the findings may not be accurate for making generalisations about other settings or populations. When this is the case the results are NOT generalisable. On the other hand, if the research uses a large cross-section of subjects, or investigates many different settings, it can be considered to be generalisable.
Representativeness
Any sample of subjects which contains a good cross-section of the population being investigated is considered to be representative (ie same proportions of males, females, different age groups, different ethnic groups etc.)
Reductionism
This refers to attempting to reduce all behaviour down to one explanation (eg. PMT and women’s moodiness).
Determinism
This refers to any explanation that suggests an individual’s behaviour is determined by factors beyond their control (eg. genes).
Ethnocentrism
This term is better understood if split into two:
- Ethno- relating to one’s own culture
- Centric- centred
- Ethno-centric – centred around one’s own culture – seeing the world from a culturally biased perspective.
Important because…
Ethnocentrism can cause a researcher to misinterpret the behaviour of their subjects. This can be due to seeing their own culture as superior and the subjects’ culture as inferior, or just because they do not understand the other culture.
Related Terms
- Eurocentric – with a European perspective (or Western perspective) – most psychology is Western (UK / USA)
- Androcentric – with a Male perspective – most psychologists are male
- Egocentric – with a self-centred perspective – most young children can not see things from other people’s perspective.
