Data Quality, Research Design, and Evaluating Research Flashcards
(46 cards)
Projective Tests purpose
Reveal the unknown or hidden
aspects of personality to the person and
researcher
Basic Method
Give someone an ambiguous
stimulus and ask them to tell you about it
Analysis
No clear cut, quantitative results
◦ Analysis of content of stories, letters, and
speeches
◦ Mostly used by clinical psychologists
◦ Provide you with B-Data
Objective Tests purpose
For the person to reveal to the researcher
what they think or know about themselves
Basic Methods
Basic Methods
Rational method
Write items that seem
directly, obviously, and rationally related to what
is to be measured (S-data
Factor analytic
Identify which items group
together using factor analysis (mainly S-data)
Empirical
Identify items based on how people
in preidentified groups respond (B-data)
Experimental Study:
Test differences between groups to determine if the difference is larger than would be expected by chance
Independent variable
a characteristic of an experiment that researchers manipulate to see if it causes a change in the dependent variable
Dependent Variable
The variable the researcher observes/ measures
Correlational Study:
Correlational method: A research
technique that establishes the relationship between
two variables by measuring both variables as they
occur naturally in a sample of participants
There are no experimental groups
* Questionnaires are administered
Comparing & Contrasting
Experimental and Correlational Methods
Both attempt to assess the relationship between two (or more)
variables
* The statistics (with two groups) are interchangeable
* The experimental method manipulates the presumed causal
variable, and the correlational method just measures it.
* Reasons for not knowing causal direction in correlational studies:
* Third-variable problem
* Unknown direction of cause (the directionality problem)
Experimental and Correlational Methods
Complications with experiments
◦ Uncertainty about what was really manipulated
A version of the third-variable problem
◦ Can create unlikely or impossible levels of a variable (“sledgehammer
manipulation”)
◦ Often require deception (ethically precarious)
◦ Not always possible
* Takeaway:
* Experiments are not always better
* An ideal research program includes investigations with both designs
Statistical significance
A result that would only occur by chance less than 5 percent
of the time
Null-hypothesis significance testing (NHST)
Determines the chance of
getting the result if nothing were really going on
p-level
Probability of obtaining a result if there is no difference between
groups or no relationship between variables (p<.05; p<.01; p<.001)
Problems with NHST
The logic is difficult to describe (and understand)
* The criterion for significance is an arbitrary rule of thumb (although the field has
agreed-upon rules of thumb)
* Nonsignificant results are sometimes misinterpreted to mean “no result” or
no relationship or difference
* Only provides information about the probability of one type of error
* Type I error vs. Type II error
Effect size
An index of the magnitude or strength of the relationship between the
variables
Correlation coefficient (Pearson’s r)
The most commonly used* measure of effect size
◦ Between -1 and +1
◦ If 0, there is no relationship
Positive correlation
as one variable goes up, so does the other; likewise, as one
variable goes down, so does the othe
Negative correlation
as one variable goes up, the other goes down
Confidence Intervals
Provides the range of values within which the true population
correlation is likely to be found
Variance
The sum of squared deviations from the mean
* The squaring is a computational convenience but has no other rationale
Not a good way to explain the magnitude of a correlation
