Exam 7: From Manipulative to Analytic Control: Quasi-Experimentation

Consider the politics of evaluation research and discuss how and why politics should be considered in evaluation research. Now imagine that your local government wants to implement a basic income of $1,000 per month for citizens earning less than $200,000 per year and they want you to evaluate the programs efficacy. Discuss the following: a. name the independent variable in this study. b. propose a possible dependent variable for this study. c. What are two rival plausible explanations? d. evaluate the study's internal validity, identify what threat(s) to internal validity you believe might be present, and state how you might safeguard the study to control for those threats. e. discuss the potential use of time-series data and why it may (or may not) be helpful to this study. f. discuss the role of politics in this process

Time-series designs should not be used in quasi-experimentation.

When we cannot randomly assign people, we should start looking for "nonequivalent" control groups to allow us to make "reasonable guesses."

Evaluation researchers should be aware of political considerations.

A dentist wants to try a new treatment plan for gingivitis but can only offer it to a certain number of people. He wants to use the remaining eligible people as a control until he can give them the treatment later. What is this control group called?

Summarize what it means to have a John Henry effect in your research and discuss how it may threaten internal validity.

Jessie would like to evaluate the academic performance of students in two different high schools. Jessie cannot randomly assign students to high schools, so how can Jessie minimize rival plausible explanations?

The three criteria of causality should not be considered when assessing quasi-experimentation.

______ is a quasi-experimental design where measurements are taken from two groups of test units. The experimental group is exposed to a treatment/independent variable, and then another series of measurements is taken from both groups.

______ occurs when some individuals drop out of the research before it's completed.

Nia is conducting a study with intact groups; thus, she needs to anticipate and address all rival plausible explanations. What type of control does Nia need?

______ is a time series of a particular outcome of interest that is used to establish an underlying trend, which is "interrupted" by an intervention at a known point in time.

______ is the view that a better world can only be created by empirically driven evaluators.

Rival plausible explanations are not a concern in traditional laboratory experimentation.

Which of the following was one of the first examples demonstrating the logic of quasi-experimentation.

Alicia is conducting a study on marathon runners by interviewing them at multiple points during the marathon, however, some individuals she interviewed leave the marathon without completing. What threat to internal validity is this?

Julie has designed an experiment that allows her to randomly assign people to one of two groups. What type of control does Julie have?

In traditional laboratory research, it is easy to isolate experimental and control groups from each other.

Single-session, short-term laboratory experimentation virtually precludes ______ as a problem. But in the field, where time-series data and a succession of follow-ups are more likely, this problem increases in relevance.