Research Methods Flashcards
(20 cards)
The role of peer review
1- allocate research funding
2- validate the quality and relevance of the research (eg conclusions drawn)
3- to suggest improvements
Evaluation of peer review
1- anonymity
2- publication bias
3- burying ground breaking research
Correlation coefficient
A number between -1 and 1 that represents the direction and strength of a relationship between co-variables
Positive - the closer the coefficient to 1 the stronger it is
Negative- the closer the coefficient to -1 the stronger it is
Used for inter observer and concurrent validity (correlation of two sets of data exceeds +80)
Case studies
Case studies - an in depth investigation, description and analysis of a single individual , group, institution or event.
- often qualitative data and longitudinal studies
Content analysis
- a research technique that enables the indirect study of behaviours by examining communications that people produce , eg texting
ASK TEACHER
Assessing reliability
Inter observer reliability- agreement between two it more observers
Test re test - assessing questionnaire or psychological test - assessing same person one two occasions
Improving reliability
Questionnaire- remove complex or ambitious questions ( replacing open q with closed q)
Interviews- using same interviewer or
Making interview structured
Observations - operationalisation of behavioural categories
Validity types
Ecological validity - findings can be generalised to other settings and situations
Population validity - research sample reflects target group
Internal validity- observed behaviour was due to independent variable
Temporal validity - can be generalised to other historical times and eras
Improving validity
Questionnaire - open qs Interviews - open qs Experiments - stratified sampling Observations- control group Case studies - compare to other similar chases
Assessing validity
Face validity - a measure scrutinised to determine whether it appears at first sight to measure what it is supposed to
Concurrent validity - extent to which a psychological measure relates to an existing similar measure
Parametric tests ( related t test, independent t test, pearsons product moment )
3 parametric assumptions must apply ..
1- data is interval
2- distribution of score are are normal ( scores cluster around the mean/ normal bell curve)
3- variances should be homogeneous ( deviation of scores is similar between conditions
Reporting psychological investigations
SUMMARY
INTRODUCTION
METHOD (procedure, design, sample,ethics, materials)
RESULTS
DISCUSSION(findings in verbal form)
REFERENCING
( author, date, title of book, place of publication, publisher)
Startisrical tests
Spearmans rho
N= no. Of participants
Statistical tests
Unrelated t- test
Df- (number of scores in group 1 + number of scores in group 2) -2
Statistical test
Related t test
Df= (no of pairs of scores) -1
Statistical tests
Pearsons r
Df = (number of participants) -2
Statistical test
Chi- squared
Df= (no of rows -1) times (no of columns -1)
Mann-Whitney
N1 and n2 = number of scores in each condition
Statistical test
Sign test
N= number of participants
Statistical test
Wilcoxon
N=no. Of pairs of scores with a difference
