Research Methods Flashcards
What is a hypothesis?
A precise testable statement that states the relationship between the IV and DV. (You can also think of it as a prediction). e.g. People with an audience will throw more balls into a bucket from 2m away than people without an audience.
What is the difference between the null hypothesis and the alternative hypothesis?
The null hypothesis is the statement of NO relationship occurring. The alternative hypothesis is when there is a relationship. (It’s annoyingly called alternative as it is the alternative to the null hypothesis but it is your actual hypothesis)
What is the dependent variable?
What is being measured. e.g. how many balls a person gets in the bucket.
What is the independent variable?
The thing being changed/manipulated. Normally this is the two different conditions of the experiment e.g. audience or no audience.
What is operationalisation?
Making the variables measurable and precise. e.g. instead of “people with an audience will throw more balls in a bucket”, you could operationalise it like this - “people with an audience of 6 people will throw more tennis balls into a 40cm diameter bucket from 2m away”
What is an extraneous variable?
A “nuisance” variable that might affect your dependent variable e.g. the colour of the room or the ball skill of the participants.
What is a standardised procedure?
Using exactly the same method and instructions for all the participants in your study.
What is randomisation?
This is simply randomising things in your experiment e.g. randomly allocating participants into the different groups or giving participants words in a random order for a memory experiment for example.
What is the difference between quantitative and qualitative data?
Quantitative = numerical data = easy to analyse lots of it Qualitative = word data e.g. transcript of an interview = harder and more time consuming to analyse but richer, deeper data
What is a lab experiment and + & - of them?
An experiment done in a lab! + Highly controlled therefore less extraneous variables - Not very true to real life, therefore low external validity
What is a field experiment and + & - of them?
An experiment done in a realistic setting. + Not so easy to control therefore more extraneous variables - True to real life, therefore high external validity
What is a natural experiment and + & - of them?
An experiment where you take advantage of (and therefore don’t manipulate or create) an independent variable that has already happened. e.g. the effects of a hurricane on stress levels. The independent variable is hurricane or no hurricane so you can take advantage of one that has happened (you wouldn’t create a hurricane yourself!) + Normally very real life situations - Don’t happen very often - can be difficult to find
What are the 3 experimental DESIGNS?
Independent groups, repeated measures, matched pairs.
What is independent groups and + & -
Split your participants into different groups for each level of the independent variable. e.g. half with audience, half without audience. + No order effects - Participant variables can be a problem (e.g. you might end up with all the basketball pros in one group!)
What is repeated measures and + & -?
All participants go through all the conditions of the experiment + Eliminates participant variables (the basketball pros do both conditions!) - Order effects - participants may be tired or bored when they have to do the experiment for the second time so the second condition might suffer. Therefore depends which order they do the conditions in.
What is matched pairs and + & -?
Pairs of participants are matched on a characteristic relevant to the experiment (e.g. ball throwing ability). One member of each pair does one condition and the other does the other condition. + Reduces participant variables - Very time consuming as you have to measure the characteristic and then do the matching before you even start the experiment.
What is counterbalancing?
This is when you are doing a repeated measures design to reduce order effects. Half the participants do condition A then condition B. The other half to B first, then A.
What is the difference between a target population and a sample?
The target population is the entire population you are looking at (and this could just be the population of the world!) e.g. effects of counselling on anxiety in the UK, the target population would be ALL anxious people in the UK. You obviously couldn’t test EVERYONE so the sample is the people who do your experiment and can be a small sample e.g. 10 or could be a large sample e.g. 1,000 (often in the case of something like a questionnaire)
What is random sampling and + & -?
Take all the names of the people in your target population, give them all a number, put all the numbers in a hat (or computer) and randomly take out the amount needed. + No bias, everyone has a chance of being selected - Time consuming to get the names of the entire target population and they might not even consent when you contact them
What is opportunity sampling and + & -?
The researcher selects people who happen to be there at the time. + Easy, cheap and quick! - May not be a representative sample of your target population. Historically most participants of psychology experiments have been university students.
What is systematic sampling and + & -?
Taking every e.g. 5th, 10th or 20th name of a list of your target population. + Avoids researcher bias. Everyone has a chance of being selected. - There is a small chance that you end up with a biased sample e.g. all men. Also time consuming getting list of entire target population.
What is stratified sampling and + & -?
Participants are selected from different sub-groups (strata) of the target population in proportion e.g. few managers, lots of factory floor workers + Very representative of the target population - Very time consuming
What are the 6 BPS (British Psychological Society) ethical guidelines for a psychological study?
Remember: What Can People Do In Psychology?
Withdrawal. Participants have the right to withdraw any time
Confidentiality. Personal details should not be accessible and anonymous if possible.
Protection from psychological and physical harm. Offer counselling afterwards if necessary.
Deception. Participants should not be severely misled. Witholding _some_ of the details of the study is considered mild deception and is usually ok.
Informed consent. Participants should give consent and be informed about the general aims of the experiment so they know what they are consenting to.
Privacy. People should be able to control any information about themselves. Their data should be kept confidential.
What is the difference between a structured and unstructured interview?
Structured: Interviewer reads list of prepared questions
Unstructured: Some questions prepared before but interviewee can be asked relevant new questions based on what they are saying
What are the strengths and weaknesses of interviews?
+ Lots of information
+ Deep insights into thoughts and feelings
- A lot of data which is tricky to analyse
- Some people are uncomfortable face to face or on the phone
In a questionnaire, what is the difference between open and closed questions?
Open - Any answer possible
Closed - Fixed answers to select from
What are the strengths and weaknesses of questionnaires?
+ Can gather information from a very large number of people
+ Easy to analyse, particularly closed questions (produce numerical data which can be analysed by a computer)
- Social desirability bias (people answer how they’d like to be seen)
- Danger of misleading questions
What is the difference between a natural and controlled observation?
Natural: Record behaviour where it would normally occur
Controlled: Researcher manipulates aspects of the environment
What is the difference between a covert and overt observation?
Covert: Participants don’t know they are being observed
Overt: Participants do know they are being observed
What is the difference between participant and non-participant observation?
Participant - researcher is part of the group being observed
Non-participant - researcher is not part of group being observed
What is interobserver reliability?
Where you have two observers, to check that your behavioural categories are clear, you get both observers to observe the same thing at the same time in a test situation and then see how their results correlate. The higher the correlation, the higher the reliability.
What are the strengths and weaknesses of observations?
+ Based on real life behaviour therefore higher validity
+ Participants act naturally if they don’t know they are being observed
- Can’t gain consent if observing in a public place
- Observer bias - the researcher might only see what they want to see to get the results they want for their experiment
What are correlations?
Correlations show the relationship or association between two things (co-variables)
What kind of graph do you use to show a correlation?
Scattergraph
What are the strengths and weaknesses of correlations?
+ A good starting point for some more in depth research
- DON’T SHOW CAUSE AND EFFECT
What do the following graphs show?
Strong positive correlation
Strong negative correlation
No correlation
What is a case study?
An in depth investigation of a single individual, event or institution
What is a longitudinal study?
One carried out over a long period of time e.g. years. Often used in case studies as the individual is followed for many years.
What are the strengths and weaknesses of case studies?
+ Best way of studying rare behaviours/situations, gain new insights into things not seen before
- Focuses on one person or thing so cannot be generalised to the whole population
What is the difference between qualitative and quantitative data?
Qualititative data is words. More in depth but more difficult to analyse.
Quantitative data is numbers. Easy to analyse but lacks depth.
What is the difference between primary and secondary data?
Primary - collected by the researcher therefore perfectly suited to the hypothesis but more timely and costly to gather
Secondary - Data from other studies or government statistics therefore easy to gather but not so perfectly suited to investigation
What is the range of a set of scores?
Range is the spread of scores and is calculated by taking the smallest from the largest. e.g.
2, 3, 4, 1, 3, 2, 3, 9
Range = 9 - 1 = 8
Can be skewed by a freak score
What is mean, median and mode?
These are measures of central tendency.
Mean - add up all the scores and divide by number of scores (affected by freak scores)
Median - middle value (not affected by freak scores)
Mode - most common score. Used when the data is in categories e.g. Strawberry, chocolate, vanilla.
What are these two charts?
The first is a histogram (continuous numerical data)
The second is a bar graph (categories, and gaps between the bars)
What does this graph show?
Normal distribution. A bell curve graph where the mean, median and mode are all the same and at the top of the graph.
