The popularity of mixed methods approaches in evaluation has increased significantly over the last decade, although there is still some controversy regarding these approaches. The mixed methods approach has gained traction in the field of policy evaluation since it represents both a compromise and a challenge for proponents of more conventional evaluation work. Mixed methods studies can range from quantitative work supplemented by case studies at one end, to qualitative work supplemented by secondary or survey types of analyses on the other. The difficulty often arises when methods employing different views of reality or methods designed for different purposes are employed in the evaluation. In addition, mixed methods may attempt to combine different approaches to bias, causality, validity, and sampling. There are also considerations having to do with levels of analyses, concurrent or sequential data collection, and triangulation.
a 2 page paper that addresses the following:
- Assume you were to employ a mixed methods approach to your Final Project. Explain how you would integrate both qualitative and quantitative methods in your design.
- Using the guidelines from the USAID (2013) resource, describe your formulation and explain your reasoning:
- What qualitative method will you use?
- How will your qualitative analysis method supplement your quantitative methods?
- Will you use triangulation?
Running head: QUANTITATIVE DESIGN 1
QUANTITATIVE DESIGN 2
Selecting Treatment and Control Groups
A field experiment occurs in a natural setting or a real-life environment of a participant. In a experimental field design, the conduct is more likely to indicate a real-life situation. The researchers impose treatment on a group or subjects of interest for instance ex-prisoners to observe the response (Gerber & Green, 2012). For the program of training the ex-prisoners, the method to be implemented when selecting the treatment will involve randomly selecting a sample from the participants who have agreed to participate in the survey. The selected sample will then be divided into three groups. The first group will be the ex-offenders, the second group will be employers who have employed ex-prisoners in the community and the last group will be professionals from a criminal justice system.
For this program, the professionals from the criminal justice system and a group of employers who have employed ex-prisoners will be the treatment group while ex-offenders will be used as the control group. The survey will then be conducted on the three groups to determine responses from the three groups. The responses from the two treatment groups will be compared to responses from the controlled group (Duke & Mallette, 2011). Treatment group responses will be used to manipulate the training services offered to the ex-prisoners to get the most out of the training. The use of random selection for both treatment and control group will ensure that bias in the field experiment is reduced and the effects of participant variables are limited.
Techniques to Address Selection Bias for Quasi-Experiment Design
Quasi-experiment design uses a comparison group that captures what the outcome would have been if the policy had not been implemented (Duke & Mallette, 2011). As a result, the policy can be seen to cause any difference in the outcome of the experiment between comparison and treatment groups. Selection bias suggests that the chance of eligible or those who volunteer to take part in the study have been systematically different from those who do not take part. Some of the techniques that might be implemented to control selection bias include regression discontinuity designs and propensity score matching.
Propensity score matching involves matching an individual on their propensity score such that the likelihood of an individual participating in the study is given by their observable characteristics. The method matches the treatment group with a similar comparison group and then calculates the difference in the indicators thus obtaining an unbiased impact estimate. In regression discontinuity designs, a given criterion is used in getting participants. A cutoff used in the selection ensures that limited resources are distributed to people who need them rather than distributing the resources randomly (Gerber & Green, 2012). For instance, in the training of ex-prisoners, more of the training services could be offered to ex-prisoner who just left the prison.
Internal Validity for Non-Experimental Design
Non-experimental design measures the variable as they happen naturally. The design has low internal validity since it does not implement the use of manipulation on a variable or control. Internal validity refers to the degree to which the design used in the study is in agreement with the conclusion, (Carter, 2005). To reduce the threats which lower internal validity while using non-experimental design, biases during the selection of the sample must be avoided. When selecting a sample to participate in the research, the participants should be selected randomly to give every member of the population an equal chance of participating. As a result, selection biases are reduced which in turn reduces bias threat to internal validity.
Since non-experimental design does not include the use of controls, incorporating a control group in the study would counter the threats due to a single-group study. For example, instead of having all the ex-prisoners participants in one group, the sample population would be divided into two groups. The first group receiving treatment while the second group to be used as the control with only program differences between the two groups. Increasing the size of the sample to be used would reduce variability in the outcome. This would counter threats due to testing increasing internal validity (Carter, 2005).
Another way of handling internal validity in non-experimental design is by researching multiple perspectives. For instance, researching how to improve the training program for ex-prisoner would involve performing surveys to various professionals from the criminal justice system, ex-offenders, and employers who have employed ex-prisons from different locations using the same survey data and then comparing responses. Besides, when conducting a study, getting enough information such as detailed information about where the study is to occur will reduce internal threats brought about by history.
Carter, . (2005). Rehabilitation Research: Principles and Applications. Elsevier.
Duke, N. K., & Mallette, M. H. (2011). Literacy research methodologies. New York: Guilford Press.
Gerber, A. S., & Green, D. P. (2012). Field experiments: Design, analysis, and interpretation. New York, N.Y: W.W. Norton & Company.
What Students Are Saying About Us.......... Customer ID: 12*** | Rating: ⭐⭐⭐⭐⭐
"Honestly, I was afraid to send my paper to you, but splendidwritings.com proved they are a trustworthy service. My essay was done in less than a day, and I received a brilliant piece. I didn’t even believe it was my essay at first 🙂 Great job, thank you!"
.......... Customer ID: 14***| Rating: ⭐⭐⭐⭐⭐
"The company has some nice prices and good content. I ordered a term paper here and got a very good one. I'll keep ordering from this website."