| Definition |
 |
|
| Term |
|
|
| Group to generalize to |
Population |
| Subset of the population. |
Sample |
| the fact researcher is interested in explaining |
Dependant Variable |
| variable used to explain the one dependent or causally prior to dependent variable. |
Independent Variable |
| When experimenter can intentionally create variation in independent variable (ie. amount of funding) |
Manipulation |
| Rival explanations for behavior of dependent variable. |
Confounding Variable |
| units observations are made (regions, municipalities, etc) |
Subjects |
| Set of data with which treatment group is compared. |
Control |
| Subjects have equal chance of being assigned to treatment group or control group-purely by chance. |
Randomization |
| Program rather than other factor caused results. |
Internal Validity |
| Can the effect be expected under similar conditions in other settings. |
External Validity |
| Convenience sampling-those accessible; volunteer sampling - volunteers-OK for descriptive results, may be bias, not representative of population |
non-probability sampling |
| Individual has equal chance of being selected for sample |
Simple random (probability sampling) |
| every Xth individual selected form list, starting randomly chosen point |
Systematic (probability sampling) |
| Population may have 2 or more groups in study-provides best results; ensures even coverage of population; maintains random selection probabilities. |
Stratified (probability sampling) |
| Used when stratified or simple random sampling would be difficult and/or expensive. |
Cluster (probability sampling) |
| As ______increases, standard error decreases |
Sample size |
HHS Shrek
