Orientation to Time
all measurements are made at some point in time
cross-sectional time orientation
a qualitative study design that documents experiences as lived and told by the subjects themselves
narrative research design
a prevalence study design that analyzes existing information about every single member of a population
population statistics
the goal of correlational research
to discover whether and how two or more variables are related to each other
the best type of an experimental design study
randomized controlled trial (RCT)
an approach that requires people to look backward over time and report what happened in the past
retrospective
a study design that involves the scientist immersing themselves as a participant in a specific social or cultural context in order to gain insight
ethnography
the difference between a crude rate and a general rate
crude = number per population size
general = number per population size among those who were eligible to experience the event
the question that mediating variables always attempt to answer
why or how does the predictor have that effect on the outcome?
how an RCT meets the requirement for causality that the temporal order be correct
RCT'S are always longitudinal, subjects are assessed by a pre-test before and after intervention with a post-test.
the reason why all experimental studies are longitudinal
they include more than one point of contact with each subject
the goal of qualitative research
to produce insight into something that is fundamentally subjective
the kind of sampling used in prevalence research designs?
no sampling involved! it includes all members of the population
the most important characteristic of the study subjects in a correlational study
as long as the sample contains variability on each of the variables measured
what are two methods to combat confirmation bias?
blinding and double-blinding
the three limitations of longitudinal data
attrition, instrument decay and practice effect
a measurement strategy typically used with narrative research design
narrative or semi-structured interviews, naturalistic observations, questionnaires
what measurement strategy is common for social survey's?
self-report questionnaires
a p-value is the probability of ______
the results have a 5% chance of error due to randomness or 95% chance the results were correct
one reason why scientists would use a quasi-experiment instead of an RCT
1) the difficulties such as money and time, wanting to find promising results first from correlational studies and quasi-experimental studies
2) variables cannot be manipulated, there's not experimental control over the intervention
3) unethical or impractical (breast-feeding, domestic violence)
the only circumstance in which a time-trend orientation is appropriate and useful
the reason why variables aren't identified in advance in qualitative designs
study subjects drive the direction of the research, variables are identified as part of the process
the two possibilities that a wide interval could mean
2) the used a smaller sample size
two methods scientists use to rule out spurious correlations
2) investigating potential mediating variables
3) replication
the problem of a quasi-experiment design that lacks a pre-test
1) we won't know if the intervention is associated with any change, not knowing state of knowledge beforehand