Discussion 2: Generalizing Study Results

Discussion 2: Generalizing Study Results

Generalizability is the extent to which research findings from your sample population can be applicable to a larger population. There are many best practices for ensuring generalizability. Two of those are making sure the sample is as much like the population as possible and making sure that the sample size is large enough to mitigate the chance of differences within the population. For this Discussion, read the case study titled “Social Work Research: Program Evaluation” and consider how the particular study results can be generalizable.

By Day 5

Post your explanation of who the sample is. Also explain steps researchers took to ensure generalizability. Be sure to discuss how the study results could possibly be generalizable. Please use the resources to support your answer.

Melissa Armstrong RE: Discussion 2 – Week 5COLLAPSE

The Video

The pre-test/post-test was another research method that contributed to the validity of the research.  Pre-test showed where the participants stood at the beginning of the research, and the post-test measured the progress that was made by the end of the research.  I do not believe that attrition was considered initially.  The social workers in the video were too confused about their next set of actions after they were forced to drop Mr. Hernandez (Laureate Education, 2013).  Attrition, only then, became a factor.  Attrition is a strong potential in a research study and must be considered in advance.

Research Method

The research method that I would recommend is a quasi-experimental design.  Quasi-experimental design is an empirical interventional study used to estimate the causal impact of an intervention on a target population without random assignment (Yegidis, Weinbach & Myers, 2018).  Funding depends on the validity of the research study.  The more effective this program is to individuals and families, the more funding the program will receive.  Due to the topic of the group and the reason for the research study, internal validity does not need to be manipulated.  Attrition affects the effectiveness of the study which is the determining factor of funding.

Measurement over Time

The study should involve measurement over time.  Measurement over time offers stability and validity to the research, making the conclusion concrete.  There would be a control group.  A control group is compared to the experimental group but does not receive the test variable.  This research method is used to find answers within the experiment.  This method accounts for the reliability of the research as well.

Research Design

In this group research design that I imagine, two groups will be measured.  This research method will measure the control group which will be the individuals that have not met the attendance requirements and then the experimental group, which will be the individuals that have successfully completed the program.  This research method will measure the effectiveness of the program in terms of reduction of undesired behaviors and attrition for the group that was partially observed.


Limitations in terms of generalizability and internal validity that I anticipate based on the research plan that I envision include many individuals that will not complete the program or not be available for post-testing.  This alters the validity of the study as well as the conclusion.


Pertaining to the client drop out, I can tell the social worker that attrition is inevitable and should be expected and accounted for early in the research process.  Individuals not completing the research study or not being available for a post-test is an often occurrence and should be expected by the researcher.  Adding attrition into the results is beneficial to the study by helping to maintain stability.

No Control Group

If there is no control group, the research design that could answer the question, “is this parent training class effective in reducing abusive parenting practices,” is the pre-experimental design.  The pre-experimental design studies a single group meaning no control group nor comparison involved in this research design (Yegidis, Weinbach & Myers, 2018).


Attrition would certainly impact the external validity of a study — would it impact the study’s internal validity as well?