9.9 You are asked to develop an experiment for a study of the effect that compensation has
on the response rates secured from personal interview subjects. This study will involve 300
people who will be assigned to one of the following conditions: 1. No compensation. 2. $1
compensation. 3. $3 compensation. A number of sensitive issues will be explored concerning
various social problems and the 300 people will be drawn from the adult population.
Describe your design.
The study presented for this question seeks to determine the impact compensation has on
response rates from personal interview subjects. By utilizing a seven step process, researchers
can conduct an experiment in an attempt to answer this question (Cooper & Schindler, 2014, p.
195).
Select Relevant Variables
Through formulation of a hypothesis, researchers identify the experiment’s variables, and
describe what impact they believe manipulation of the independent variable has on the dependent
variable (Cooper & Schindler, 2014). For this experiment, the hypothesis is that the more
individuals are compensated for their participation in a personal interview, the more likely they
are to respond. In this scenario, the independent variable is the compensation, with the dependent
variable being the response rate. Thus, the purpose of the experiment is to determine whether
manipulating compensation has an impact on participant response rate.
Specify the Treatment Levels
Cooper and Schindler (2014) define treatment levels as “arbitrary or natural groups the
researcher makes within the independent variable of an experiment” (p. 195). The number of
levels will vary based on the hypothesis, but Collins, Dziak and Li explain that often times
experimenters are limited to fewer levels than they would like, due to logistical difficulty and
expense (2009). In this experiment, there are three levels of the independent variable: 1. no
compensation; 2. $1 compensation; 3. $3 compensation. The group receiving no compensation
acts as the control group, providing a base level for comparison (Cooper & Schindler, 2014,
p.197).
Controlling the Experimental Environment
To ensure the accuracy of the experiment’s results, all aspects of the experimental
environment are held constant for all participants, in each of the three groups (Cooper &
Schindler, 2014). Additionally, Choi and Lam (2017) explain that experimental outcomes may be
unintentionally biased by human expectation, particularly if the outcome is subjective. For this
reason, a double blind study, in which neither the participant nor the experimenter know which
group the individual is assigned to, is considered to be the most effective experimental design
(Choi & Lam, 2017). To make this experiment double blind, the interviewer should not know
whether, nor how much, the participant has been compensated; and the participant should not be
made aware of the fact that other participants were compensated more, less, or not at all.
Choosing the Experimental Design
For this particular experiment, the posttest-only control group design is the most effective
experimental design. Morris (2008) explains that a pretest-posttest approach enables researchers
to control for preexisting differences across participants. In this experiment however, participants
are being randomly assigned to one of three groups depending solely upon the amount of
compensation they have been offered, and therefore a pretest is not necessary (Cooper &
Schindler, 2014).
Selecting and Assigning Participants
Participants for this experiment are solicited via an online posting. Respondents are then
screened to ensure they meet the age requirement, as the experiment is open only to adults.
Differences in age, gender, and other socioeconomic factors will be accounted for via
randomization. Random assignment ensures that the groups are as comparable as possible
(Cooper & Schindler, 2014).
Because the interview is said to involve a number of sensitive issues, participants’ right to
privacy should be respected via confidentiality agreements, nondisclosure documents, as well as
restricting access to participants’ identifying information (Cooper & Schindler, 2014, p. 34).
Pilot Test, Revise, and Test
The purpose of this step is to reveal errors in the experimental design, as well as identify
extraneous or environmental conditions which have the potential to impact the research design or
skew results (Cooper & Schindler, 2014). An example of an extraneous or environmental
condition that could impact the effectiveness of the experiment would be the website utilized to
solicit participants crashing, preventing individuals to sign up for the experiment.
Analyzing the Data
In this final phase, data from the experiment is analyzed to ensure validity and determine
if the results support or disprove the hypothesis.
Biblical Integration
The experiment in this example is said to explore sensitive issues concerning various
social problems, topics which may be difficult for research participants to discuss with a total
stranger. As Christians, we are called to be empathetic and show compassion for one another.
There is no greater example of a compassionate being than Jesus Christ himself, who died on the
cross for our sins. Colossians 3:12 says, “Therefore, as God’s chosen people, holy and dearly
loved, clothe yourselves with compassion, kindness, humility, gentleness and patience” (NIV).
References
Choi, S.W., Lam, D.M.H. (2017). Credible control groups and blinding. Anaesthesia, 72(3), 402-
404.
Collins, L.M., Dziak, J.J., Li, R. (2009). Design of experiments with multiple independent
variables: a resource management perspective on complete and reduced factorial designs.
Psychological Methods, 14(3), 202-224.
Morris, S.B. (2008). Estimating effect sizes from pretest-posttest-control group designs.
Organizational Research Methods, 11(2), 364-386.
Powered by TCPDF (www.tcpdf.org)
