Term
| Threat to internal validity/history |
|
Definition
| An event occurring between pretest and posttest other than the independent variable that could effect the dependent variable.Best way to control this is by using a control group that receives no treatment. |
|
|
Term
| Threat to internal validity/maturation |
|
Definition
| Changes occurring in participants because of passage of time. A control group will eliminate this threat. |
|
|
Term
| Threat to internal validity/testing |
|
Definition
| Taking a pretest somehow affects the taking of the posttest. Use of a control group that does not receive any pretesting will reduce this threat to internal validity. |
|
|
Term
| Threat to internal validity/instrumentation |
|
Definition
| Instruments are not accurate/precise enough or do not measure what they are supposed to measure. |
|
|
Term
| Threat to internal validity/selection bias |
|
Definition
| Participants selected in a non-random manner differ in some way. Recruiting volunteers and then randomly assigning them to groups is better than allowing the volunteers to self-select a group. Matching participants on selected characteristics and then randomly assigning them to groups can also reduce selection bias.Pretesting groups on measures of the dependent variable to make sure there are no pretreatment differences between groups is another control that can be used. |
|
|
Term
| Threat to internal validity/selection maturation |
|
Definition
| Using intact groups that vary in some element of maturity. Pretesting and/or prescreening groups is a way to avoid this threat. |
|
|
Term
| Threat to internal validity/statistical regression |
|
Definition
| Extremely high or extremely low scores regress towards mean. To counter this weakness a study could be designed to follow a random sample rather than in high low groups based on one testing. |
|
|
Term
| Threat to internal validity/mortality-attrition |
|
Definition
| Participants drop out of the study or cannot be located. Using incentives to encourage participants to stay in the study may be helpful i reducing mortality. |
|
|
Term
| Threat to internal validity/Hawthorne effect |
|
Definition
| Altered behaviour due to the effects of being studied and observed. To control for this , a researcher may try to prvide the control group with some type of special treatment that is comparable to the experimental group but would not have a direct impact on the dependent variable. Therefore, with both groups receiving special attention, any differencesin the dependent variable would likely be from the intervention. Another way to try to compensate for the Hawthorne effect is to keep participants from knowing that they are taking part in a study or being observed. |
|
|
Term
| Threat to internal validity/placebo effect |
|
Definition
| altered behaviour because of expectations. To control for this the researcher tries to make sure that those in both the control and experimental groups receive the same information so that both groups would have similar expectations. If both groups have similar expectations at the begining of the study, any differences between groups are more likely from the treatment effects than placebo effects. |
|
|
Term
| Threat to internal validity/diffusion of treatment |
|
Definition
| Treatment of experimental group spills over to comparison or control groups.Controlling for a diffusion effect can be difficult, particularly if those in the experimental and control groups live and work together. |
|
|
Term
| Threat to internal validity/location |
|
Definition
| This occurs when ther eare differences in locations where interventions take place. The best method to control for a location effect is to make the locations the same for all participants. |
|
|
Term
| Threat to internal validity/implementation |
|
Definition
| This involves the individual or individuals responsible for implementing the experimental treatment and the possibility that they may inadvertently introduce inequality or bias into the study. To control for this have someone other than the program developer present the program. |
|
|
Term
| Threat to external validity/selection treatment interaction |
|
Definition
| This conserns the ability of a researcher to generalize the results of a study beyond the groups involvled in the study. Research sophistication and difficulty increase as on desires to generalize to larger groups.Carefully consider what groups one can generalize to, and do not generlize to outside groups. |
|
|
Term
| Threat to external validity/setting treatment interaction |
|
Definition
| This concerns the extent to which the enviromental conditions or setting under which an experimental study was conducted can be duplicated in other settings.To contrl this one can always duplicate setting factors as nearly as possible when replicating a program with a different population. |
|
|
Term
| Threat to external validity/history treatment interaction |
|
Definition
| This threat develops when the researcher tries to generalize findings to past and future situations.To control for this one can always duplicate historical factors as nearly as possible when replicating a program with a different population. |
|
|
