Lecture 2 Flashcards
Conceptualization
Identify the concepts you want to study. (e.g., empathy, prosocial behavior, violence)
Operationalization
Specify how you define or measure the concepts in your research question. (e.g., How do you measure empathy?)
Dependent variable
A variable that is hypothesized to vary depending on the influence of another variable (i.e., what you measure/observe)
Independent variable
A variable that is hypothesized to cause or lead to changes in another variable (i.e., what you define, control or manipulate, so that you can measure the effect on the dependent variable).
non-manipulable
Note that many variables in social research are (e.g., gender, ethnicity, socioeconomic status).
Nominal level
no mathematical interpretation; categories vary in quality. e.g. omnivore, vegetarian, vegan, or fruitarian
Ordinal level
categories have a logical order. e.g. fail, pass, or distinction
discrete measures
Measures at the nominal and ordinal level are also called
Interval level
equal intervals represent equal differences. e.g. temperature, year
Ratio level
contains absolute zero e.g. reaction time, distance
continuous measures
Measures at the interval and ratio level are also called
validity
indicates whether conclusions are well-founded.
Internal validity
Are the causal relations between variables real?
Causality
A concern with establishing a “cause and effect” connection between variables, rather than the mere relationship between them
Confounding variables
variables beyond the operationalized dependent or independent variables that could influence the findings.
Experimenter bias
the behavior or actions of the experimenter may influence the responses of the participants or the data collection in general
Single blind experiment
Information that could bias the results is withheld from the participants.
Double-blind experiment
Information that could bias the results is withheld from both the participants and the experimenter.
External validity
Can results be generalized to other settings or other populations
Reactivity
Participants adjust their behavior/responses because they know that they are being observed
Demand characteristics
Participants adjust their behavior/responses according to what they believe the researcher expects or hypothesizes.
Ecological validity
Are findings applicable to everyday life? Does the research setting resemble a “real-world” situation
Reliability
Are measures consistent? (i.e., do they give the same results over time when the phenomeon has not changed?)
Measurement validity
Type of validity associated with whether an indicator really measures a concept
Face validity
does the measure reflect the concept in question “at face value
Content validity
Does the measure cover the full range of the concept’s meaning? (e.g., does the measure cover different aspects of scientific curiosity?).
Criterion validity
Compare scores on the newly developed measure with scores on another more direct or already validated outcome measure of the same phenomenon
Concurrent validity
Compare the scores on the measure with another outcome measured at the same time (e.g., is the measure positively related to the extent to which people choose to read short stories about scientific discoveries?
Predictive validity
Can the measure predict a future outcome? (e.g., can the measure predict who often reads popular science books/magazines or visits popular science websites?)
Construct validity
showing that a measure corresponds with established measures of theoretically related concepts. (e.g., is the scientific curiosity scale positively related to a validated general curiosity questionnaire?)
Discriminant validity
showing that a measure is not related (or negatively related) to measures of concepts that should theoretically not be related
Triangulation
the use of two or more methods to study the same research question, or the use of two ore more measures to indicate the same variable.
Convergence
between measures increases confidence in the validity of the measures
Divergence
between measures could indicate measurement error or that the measures genuinely tap different concepts
Cross-sectional design
- A study, measuring more than one case at a single point in time to test variation (often a survey/ questionnaire)
- Quantitative data
- Examine relationships between variables (i.e., correlation)
- no conclusions about causality (= weak internal validity)!
Longitudinal design
- Research in which data is collected at two or more points in time
- Very intensive and expensive in terms of time and resources
- Potential issues with attrition and subject fatigue
Panel study
cross-sectional sample that is followed over time
Cohort study
sample consists of a group that experiences some event (such as being born) in a selected time period
Case study design
- Detailed and extensive analysis of one case.
- Often qualitative, but can also be quantitative
Comparative design
- Comparing two or more cases, or two or more samples.
- Often quantitative in survey form, but can also be qualitative