terms from lectures Flashcards
grounded theory
An approach to the analysis of qualitative data in which the goal is to use the data to generate theory; the data collection and analysis proceed in an iterative (recursive) fashion.
middle range theory
limited in scope and can be tested using empirical evidence.
grand theory
general and abstract. provide ways of looking at the world that can be an inspiration for research.
3 categories of epistemology
- positivism. (empirical, testable, science = value free.)
- interpretivism. (critique of positivism, focuses on subjective meaning.)
- critical approach. (also critique of positivism, science =/= value free.)
ontology
branch of metaphysics dealing with study of being.
epistemology
study of knowledge and how it is acquired.
2 ontological perspectives
- objective. (reality is objective, outside of perception.)
- constructionist. (reality is set of mental constructions.)
reflexivity
researcher’s awareness that values, decisions, biases, and presence impact the research. want to make bias explicit; reflexive on process.
response set
A term for the tendency for some people, when responding to multiple-indicator measures, to respond to every item in the same way, suggesting that their answers are motivated by something other than their actual feelings. Three of the most common response-set effects are “acquiescence,” “social desirability,” and “laziness or boredom.” impossible to completely avoid.
probing
used when interviewer needs more information from respondent. can be problematic so should be standardized to avoid influencing answers.
prompting
when interviewer suggests a specific response to the interviewee. rare. last resort.
researcher driven diaries
alternative to using questionnaires and interviews, especially when interested in behaviours rather than attitudes. participants record feelings, perceptions, and actions shortly after they occur.
ecological fallacy
characteristics of the group are falsely applied to individuals.
2 research design explanation types
- nomothetic. (cause and effect, expressed in general laws/principles. quantitative.)
- idiographic. (rich descriptions, qualitative.)
measurement/construct validity
measuring what you purport to be measuring.
2 types of longitudinal design
- panel. (same group studied at diff times.)
- cohort. (diff groups studied at diff times. less attrition.)
panel conditioning
when people’s actions or beliefs change after participating in panel.
3 types of case
- critical case. (shows conditions under which hypothesis holds/does not hold.)
- extreme/unique case. (study extreme to understand common.)
- revelatory case. (case/context never before studied.)
2 types of concept definitions
- nominal. (describe concept in words.)
- operational. (describe how to measure concept.)
codes
labels or titles given to themes or categories.
internal reliability/consistency
whether multiple measures administered in one setting are consistent.
concurrent validity
measure correlates with criterion thought to be relevant to concept.
construct validity
concepts relate to each other in a way consistent with researcher’s theory.
convergent validity
one measure of concept correlates with second measure of same concept using different measurement technique.
kendall’s tau-b
shows correlation between pairs of ordinal variables, or one ordinal and one interval-ratio variable.
spearman’s rho
correlation between pairs of ordinal variables.
cramér’s v
shows strength of relationship between two nominal variables.
chi-squared
measures likelihood that relationship between two variables exists in population.
spuriousness
exists if two variables are correlated but only through a third variable. two IV’s.
systematic sample
type of probability sample. participants selected directly from sampling frame without using random numbers.
periodicity
problem with systematic sampling that occurs if cases in sampling frame are arranged in some systematic order. increases chances of sampling error.
multi-stage cluster sampling
type of probability sampling for large populations with no adequate sampling frame. 2 or more stages of picking clusters. problem of clusters being different sizes and cluster samples being stratified as well.