TOPIC 5:
Good Study Design
| Objective 8: Describe the elements of a good design (controlling the environment, controlling equivalence of subjects and groups, controlling the treatment, controlling measurement, and controlling extraneous variables). |
| What are elements of good study design? |
- Controlling the environment is a major aspect of study control because unexpected occurrences in the environment introduce unidentified and unplanned for extraneous variables. Consistency in the environment for all subjects is critical. When reading about the study setting, consider factors that may influence the relationship between the IV and DV.
- Controlling equivalence of subjects and groups requires different strategies.
- Subjects in a study need to be as similar as possible for all characteristics except the variable(s) of interest. This provides control so differences in the variables are not because of uncontrolled differences in the subjects. The sampling criteria, inclusion and exclusion, specify who can be in the study and provide a greater chance of obtaining a homogeneous sample.
- Groups in a study should be as different as possible on the variable(s) of interest. This may sound confusing, but, when you are thinking about the group differences, keep in mind that this difference is in the relationship between the IV and DV. In other words, you would hope your treatment/ intervention is strong enough to make a difference in the outcome variable (DV). So, the group difference(s) lie in who receives intervention/ treatment and who serves as the control group that receives no intervention/ treatment.
- Controlling the treatment means that the intervention/ treatment is administered to all subjects exactly the same way. The level of control in some nursing studies may not be so complete. For example, in hospital-based studies multiple nurses may be involved in delivering the intervention and measuring the DVs after the intervention. Some of the control of the treatment is relinquished to meet the realities of collecting data on subjects each day—all day. The level of control of the treatment differs with the environment.
- Controlling measurement means the measures used in a study should be reliable, consistently produce the same results, and should be valid, measuring what they are intended to measure. Control of measurement also means that the measures are presented to all subjects the same way with the same instructions.
- Controlling extraneous variables helps to control for rival
hypotheses of differences between the IV and the DV, other than the
intervention/ treatment. Possible extraneous variables should be
considered when designing a study.