Research Methods Flashcards
Directional Hypothesis. One or two tail?
Explicitly states the direction of difference/relationship
One-tail
Non-Directional Hypothesis. One or two tail?
Does not state the direction of difference/relationship
Two-tail
Independent Variable
Aspect of experiment which is manipulated by the researchers so the effect on the DV can be measured
Dependent Variable
Measured variable, changed caused by changed in IV.
Operationalisation
Clearly defining variables in terms of how they can be measured
Extraneous Variable
Any variable that is not taken into account which impacts the DV and is not controlled
Confounding Variables
Any variable, either than IV, that has an impact on the DV yet we cannot be sure how much it did. Means the source of the DV’s change is can not be truly identified.
What is standardisation?
What are its effects?
Ensuring that all participants have the same experience within the experiment. This is done by making sure the procedure and instructions are identical.
Prevents experimenter bias and limits confounding experiments. Makes it easier to repeat/make reliable
What is the concept of significance?
Just because a difference is found, does not mean that it is significantly different. You have to check that the results are not just due to chance. You can work this out via statistical testing
What are null and alternative hypothesis?
Null: States the opposite of the alternate. Called it because it says it will not.
Alternate: States the expectation of the experiment’s outcome.
What’s opportunity sampling?
Example?
Choosing ppts based on who’s in a certain location. Has to be relevant/appropriate.
Experiment on effects of energy drink using ppts from a gym
What’s random sampling?
When there is no strategy involved, in a school you get a register and randomly choose names.
What’s stratified sampling?
Groups of ppts are selected in proportion to their frequency in target population
What are the 2 types of error?
Type 1: Alternate accepted, Null rejected, yet should be the other way around. A sig. difference does not exist
Type 2: Null accepted, Alternate rejected, yet should be the opposite. A lack of significance is perceived to have been found
What is the concept of significance in psychological research? Or why do we use tests of significance?
Just because a difference is found, does not mean that it is significantly different. You have to check that the results are not just due to chance. You can work this out via statistical testing
What are the calculated and critical values?
Calc: The results of the test done. Has to be compared to Crit. Value
Crit: The amount that calculated has to be compared to. Formed differently based on test used.
What is the Rule of R?
If test has an R involved, then the calculated value has to be GREATER than the critical value to be significant
If test does not have an R involved, then the calculated value has to be LESS than the critical value to be significant
What are the conditions for the Mann-Whitney test?
- Independent Groups ONLY
- Both ordinal or interval data
What are parametric tests?
- These produce high level, accurate data. This means it is only one type of data - interval.
- This is because the units have to be exact. Because they are so accurate, it assumes normal distribution of ppts. (bell curve)
- Cannot be used if sample size is small (below 100). This is because it would prevent normal distribution.
- Therefore stratified sampling is great for this type.
What are the conditions for the Unrelated t Test?
- Interval data only
- Only independent group designs work because their results are not related to each other
- Homogeneity of variance: Variance in ppts/population is equal
What is Homogeneity of variance?
Variance in ppts/population is equal
What are the conditions for Related t Test?
- Has to be repeated measures as the ppts do the test twice - therefore results related!
- Measurements are interval data, typically involve ppts of normal distribution and homogeneity of variance is assumed due to the design
What about spearman’s rho?
- A nonparametric test of correlation between 2 values
- Ordinal or interval data.
- Design type doesn’t matter as long as it is correlational rather than experimental
What about pearsons r?
- A parametric test
- Interval data.
- Design type doesn’t matter as long as it is correlational rather than experimental
How about Chi-squared?
- A test of difference or association. Independent measures and nominal data ONLY!
- To calculate the degree of freedom you (rows-1) x (column-1). This is your critical value.
What is the role of peer review?
Before research can be published it must be subject to peer review by a small group of experts within the field. Such a review needs to be objective and impartial.
What are the aims of peer review?
- To allocate research funding.
- To validate the quality and relevance of research
- To suggest amendments or improvements.
What is standard deviation?
- How far the results are spread from the mean.
- This can only be done with normal distribution
- Therefore it excludes extreme data and may skew our understanding
What is a case study?
- An in-depth analysis of an individual, group or event which are considered unusual.
- Yet these can also look at more ‘typical’ occurrences such as an elderly person’s ability to recall their childhood
- Typically these produce qualitative data as a case history is developed through interviews, observations, questionnaires or a combo of all 3!
- Usually a longitudinal study and also involves collecting data from relatives and friends as well as the ppt
Strengths of a case study?
- Offer rich, detailed insight information and can shed light on rare scenarios.
- Yet they can also provide info on general functioning
Limitations of a case study?
- Small sample sizes mean it’s hard to generalise onto population
- There may be low validity as the reporter chooses the data that goes in the report and accounts may be inaccurate
