Process Of Research Flashcards
Variables
- What we measure
- Phenomena that change and can be measured
- Between individual over time or different individuals
- can be compared in many ways
Categorical and measured variables
Categorical: categories like SSPS, (e.g. material status)
Numerical: numbers like SSPS, (e.g. degree of agreement)
E.g.
the political party people vote for
Vs
Degrees of agreement with political acts
Measuring variables
Easy:
height, weight, age
Difficult:
Characteristics
Definition of variables
To be able to replicate 1:1
Directly measurable: anxiety=biting lip
Dispositional characteristics: anxiety= individual diff of how people behave
Hypothetical construct
Phenomenon is assumed to exist
While assumed on observations the phenomenon might not directly be observable
Reliability and validity
Reliability: constant measurements
Validity: do you measure what you want to measure
Samples
Sampling frame: population
Accessible population: who can be sampled
Sample: actual participants
Inclusion criteria
Sampling bias
Over or underrepresenting a group,
-convenience sample, taking people around you not representative of target population
representative samples:
-epsem (equal prob. Selection method)
-simple random sample
-systematic random sample
-stratified sampling: using procentages of population in research too
Causality
Not all data has causality yet important
-might have a relationship but A doesn’t cause B
-third factor C could be at play
Design
Cross sectional
Longitudinal
Randomized
Cross sectional: variables at the same time point measured (e.g. you measure opinions on climate change of first year uni students and how A affects B)
Longitudinal : variables measured on two or more time points
Randomized: select random ppl
Independent and dependent variable
Independent: manipulated variable
Dependent: measured
Design
Recourses (funding etc)
Research aim, sample characteristic
Previous reseatch
Hypotheses
Doesn’t have to be true
Based on past research
->Direction of relationship stated
Directional causal: more a causes more b
Directional non causal: more a is related to more b
Non directional causal
Non directional non causal
Covariance
Average amount that the data varies from the mean
Of variables are related then the changes should be similar
Standardization shows correlation
Statistical hypothesis testing
H0 hypothesis = no relationship
HA hypothesis = relationship