Comparing Research Strategies Flashcards
What are the types of research designs?
- descriptive research
- correlational research
- causality and explanation
- experimental strategies
- quasi-experimental and non-experimental strategies
- qualitative and mixed methods
What is descriptive research?
Most simple design - describing how things are!
Uses observational and survey research.
e.g. RQ: What is the typical number of hours spent studying each week?
High ecological validity: because you’re not influencing anything / not putting them into a lab, you are not influencing their actual behaviour.
Outline correlational research
Goal: describe the strength and direction of relationships between 2+ variables
- extent to which the score on one variable predicts the score on another variable
- no manipulation
Relationship: positive / negative, linear / nonlinear, monotonic / non-monotonic (2+ directions)
Outline pearson’s correlation
Gives us p and r values:
p = probability that our results could happen on accident (therefore we want our p < .05)
r = ‘Pearson’s product moment correlation coefficient’ - the strength and direction of the relationship
But: when the line is flat, r = 0 (there is no relationship)
e.g. a weak correlation (r = .2) may be statistically significant, but account for very little variance.
What is r^2
R^2 is the proportion of variance in one variable accounted for by the other variable (can only be positive)
What are the cut offs for r?
Small: r = 0.1
Medium r = 0.3
Large r = 0.5
Difference between correlation and causation?
Correlation does not equal causation
- directionality problem (A cause B or B cause A)
- third variable problem (does C cause B and A)
What are the strengths of correlational research?
- 4th year design
- preliminary work for experimental research
- practical and ethical benefits
- ecological / external validity due to no manipulation
- can use to negate theories
Correlational hypotheses example:
scores on the X scale of emotional intelligence were expected to be negatively correlated with the number of panic attacks reported.
What is an experimental research strategy?
identifying cause and effect relationships via
- manipulation of IV
- measurement of DV
- control of extraneous variables
- comparison of groups
What is a within subjects design?
Same participants in each condition
+ Not vulnerable to individual difference threats to validity
- Vulnerable to time related threats to validity
What is a between subjects design?
Different participants in each condition
- vulnerable to individual difference threats to validity
+ not vulnerable to time related threats to validity
What’s the difference between extraneous variables and confounds?
Extraneous variables: all variables in a study other than IV and DV
Confounding variables: extraneous variables that change systematically along with the IV and can influence the DV
How to control for confounding variables?
Identify potential confounds by reviewing literature.
Hold variables constant (time of day, gender, education level, etc) or restrict the range.
Match values through block randomising –> balancing levels of the variable across treatment conditions
Randomisation outline:
Use of random process to avoid a systematic relationship between variables
Does not guarantee that there will not be differences between groups - just that they are not systematic.
