Week 7 Flashcards
Used to examine the effects of exposure to a variable of interest that is not under the control of the researcher
Observational research
Start by first identifying the determinant or risk factor and then follow the participants overtime until they develop the outcome
Prospective research design
Start by first identifying the outcome and then looking backwards in time to see if they can find the determinant or risk factor that contributed to the outcome
Retrospective research design
Used when researchers initially suspect there is an association between exposure and outcome at the population, community, or country level
Ecological (ecologic) research design
Often used to measure the association in an ecological research study, despite existing literature questioning the appropriateness of the statistical procedure
Correlational analysis
The assumption that the relationship that exists for groups is assumed to also be true for individuals
Ecological fallacy
Used when a researcher is determining simultaneously the exposure and outcome for everyone in the study
Cross-sectional research design
Related to both the exposure variable and the outcome variable, causing a spurious relationship
Confounding variables
In which the relationship between the exposure variable and outcome variable is due to a third variable, which is referred to as a confounding variable
Spurious relationship
The number of cases in a population in a specific time period, expressed as the proportion of the total population at risk for the condition
Prevalence
Both individuals who have and who have not been exposed or followed overtime until the outcome of interest occurs
Cohort research design
The occurrence of new cases of the outcome
Incidents
The number of new cases arising in a given period in a specified population and is usually expressed as number of cases per 1000
Incident rate
Used to measure the association in a cohort research study design
Relative risk (RR)
Individuals are selected based on the outcome and are compared to individuals who do not have the outcome (controls), according to past history of exposure to a disease, toxin, or an event
Case control research design
Refers to the selection of persons to include to serve as controls that are similar to the cases
Matching
Used to measure the association in a case control research design study
Odds ratio (OR)
Refers to the extent to which the researcher can generalize the studies findings beyond those persons included in the study
External validity
A type of observational research in which respondents are asked a series of questions
Survey research
Occurs when researchers survey a sample of the population instead of conducting a census
Sampling error
Involves collecting data from everyone in the population
Census
Results in chance variations in individual responses around the actual or true values of what is observed for the group
Random error
Also called systematic bias, is consistent, repeatable error, associated with faulty measurement instruments or flaw procedures associated with the data collection
Systematic error
Occurs when researchers do not have a complete list of names of the individuals in the sampling frame from which the sample is drawn
Coverage error
Error associated with the observation itself
Measurement error
Occurs when all prospective respondents did not respond to the survey
Non-response error
Administering a survey only at one point in time. Does not allow you to assess changes in the construct of interest.
Cross-sectional survey design
Administering a survey at multiple points in time. Allows you to assess changes in the construct of interest.
Longitudinal survey design
Require the respondents to respond to the question in their own words
Open ended questions
Have a list of responses from which the respondents choose their response
Close ended questions
Sometimes referred to as “ two in one questions” as they ask multiple questions in a single question
Double barreled questions
The tendency to give socially acceptable responses
Social desirability
The tendency of persons to respond in a positive manner to a question regardless of the question being asked
Acquiescent response bias
In which the individual does not receive assistance from others when completing the survey
Self administered survey
Ones that are mailed to the prospective respondents
Mailed surveys
An effective way of increasing the response rate and usually sent within 2 to 3 weeks after the original survey was sent
Follow up mailing
Ones in which prospective respondents are interviewed by a trained interviewer over the telephone
Telephone surveys
A method of selecting individuals to participate in the telephone survey by generating telephone numbers at random
Random digit dialing (RDD)
Conducted via email or websites
Online surveys
Calculated by dividing the number of completed surveys by the number of surveys that were sent
Response rate
Occurs when there is a significant difference between those who completed the survey and those who did not
Nonresponse bias
