Final lecture - review Flashcards
How are qualitative and quantitative research different?
In Quantitative Research the emphasis is on quantifying: measuring, statistics, surveys and experiments. In Qualitative Research the emphasis is on understanding: gaining an insider perspective. Qualitative analysis such as thematic analysis are used, and data is gathered through interviews, focus groups, etc.
What is the research cycle?
Problem - Model and hypothesis - research design - measurement - data collection - data analysis - generalisation - problem …
Explain deduction vs induction.
deduction is from the general to specific, while induction works from specific to general.
What is falsification?
We cannot prove theories to be correct we can only prove them to be wrong.
What is a variable?
Anything that varies (age, self-esteem, etc.)
What are the four types of variables?
Nominal - groups (gender)
Ordinal - ordered groups (spiciness)
Interval - equal distance between points (temperature)
Scale - equal distance between points AND 0 has interpretation (age)
Explain independent vs dependent variables?
Independent variables are the cause/predictor/manipulation(in experiments), Dependent variables are the effect/outcome/the thing that is predicted.
What are the types of validity?
Internal validity - are the conclusions valid
External validity - Can results be generalised
Ecological validity - can results be generalised to the real world
What is reliability?
refers to consistency, can the results be replicated?
What are the two types of study designs?
Correlational designs
Experimental designs
What are the types of central tendency?
Mode, Median, Mean.
What are the types of dispersion?
Interquartile range and standard deviation
What are the types of frequency distributions?
normal distributions, bimodal/multimodal distributions, skew and kurtosis, boxplots.
What are the main steps a researcher will go through when using thematic analysis?
- familiarising yourself with the data and identifying potential interest
- generating initial codes
- searching for themes and subthemes
- reviewing potential themes
- defining and naming themes
- producing the report
What are the types of measure?
behavioural measurement, self-report measures, physiological measurement.
What are latent variables?
Variables that cannot be directly observed.
What is the formula for the Classical test theory?
Item score = true score + error
What are the measurement problems particularly with surveys?
Acquiesence bias, consistency bias, and social desirability bias.
In regards to measurement problems what are the types of validity and reliability which you should consider in developing a scale?
- content validity
- convergent validity
- discriminant validity
- concurrent validity
- Internal reliability
- test-retest reliability
What key issues are included in the BPS guidelines for research with human participants?
Consent, Coercion, Deception, debrief, withdrawal, anonymity and confidentiality, protection of participants, giving advice.
What are expectancy effects?
unconscious influences that an experimenter has on study results.
What are confounds?
Variables other than your IV that can impact your results.
What are confidence intervals?
Because we work with samples and not entire populations, there will always be uncertainties about whether our sample mean represents the true population mean. We can see this uncertainty through confidence Interval is a range of values in which the true population mean will be.
How do we calculate confidence intervals?
Mean - 1.96 x S.E. = lower bound of CI
Mean + 1.96 x S.E. = upper bound of CI
Two means are significantly different if…
One mean falls outside the other mean’s confidence interval.
What are the two types of effect size?
Pearson’s r = effect size for the strength of association
Cohen’s d = effect size from mean differences (when you compare two means)
What are the tests for between-subject designs and within-subject designs?
between-subject design - Independent-samples t-test(P) and Mann-Whitney U test (NP)
within-subject design - paired sample t-test(P) Wilcoxon signed-rank test(NP)