Research Methods in Clinical Flashcards
Type I v Type II Error
Type I Error - Rejects a null hypothesis when it is true
Type II Error - Accepting a null hypothesis that is not true
Skewed Data
Mean at Bottom, Median in Middle, Mode at Top
Mean Median and Mode are not all in the middle in Skewed distribution (like they would be in normal distribution)
Positive Skew - Most scores bunched around the Left
Negative Skew - Scores bunched around the right
Primary Data
Collected by researcher specifically for the current study.
Cons
- Takes more time to collect than secondary
- May be better to adjust the aims and use 2ndary data
Secondary Data
Data collected for a previous study
Cons
- May not quite fit the needs of the current study
- May waste time if it is found that the data may not be relevant to the current study’s aims
Longitudinal Study
Research that takes place over time so that the effects of time / age can be studied
Pros
- Allows for tests of age to be tested
- Ppt. variables controlled as is repeated measures
Cons
- Cohort effects may occur as the group studied may all be of similar age or experiences, thus not generalisable
- Attrition (dropping out). The issue is that some types of people more likely to drop out than others e.g. those with personal problems. This reduces generalisability
Cross-Sectional Research
Compares behaviour of different sections of the population e.g. ages
Pros
- May be a better test of time than longitudinal as takes less time
Cons
- Ppt. variables not controlled
- Cohort effects due to differences of the groups studied
Cross-Cultural Research
Compares behaviour in different cultures
Pros
- Can investigate nature v nurture
- Gives contrasting info as a lot of psychological studies are based in America
Cons
- Most are done by western psychologists, which could lead to researcher effects as they have expectations about the differences
- Tests used may not be valid in other cultures e.g. IQ Tests
Meta-analysis
Results of several studies
Pros
- Increases sample size
- More generalisable as many different groups can be studied
Cons
- Selection bias, may want to pick studies to use that have more positive results
- May have used different methods, so not appropriate to combine the data