PART TWO: Spot the Internal Validity Threat that is MOST
likely at work for each of the following scenarios: Use the key
below to match the study validity description with the specific
validity threat
Maturation
Regression to the mean
Selection
Mortality
Instrumentation
Testing
History
Interactions
Diffusion
No Threat
The study ran for several weeks during the semester. About a
week after it started, the university announced that they would be
holding training sessions for faculty and staff about how to handle
situations involving a gunman on campus. This shut down the study
for several days as the university needed the lab building for
training. The study then resumed according to script after the
training. The researchers found that those in the experimental
group did not differ in their memories regarding the presence of a
gun compared to those in the control condition. That is, the mean
score that a gun was present was similar for the experimental group
(M = 65%, SD = 11.4%) and the control group
(M = 63%, SD = 13.26%).
Maturation
Regression to the mean
Selection
Mortality
Instrumentation
Testing
History
Interactions
Diffusion
No Threat
The experiment proceeded as normal, with the experimental group
hearing about the potential presence of a gun as the confederates
left the study and the control group not hearing about the gun. All
participants returned on day two for the interview with the police
detective. After debriefing, none of the participants said they
were aware that the theft was staged. The researchers found that
those in the experimental group did incorporate the gun
misinformation into their memories of the theft more than those in
the control condition. That is, the mean score that a gun was
present was much higher for the experimental group (M =
67%, SD = 12.4%) than the control group (M = 33%,
SD = 11.16%).
Maturation
Regression to the mean
Selection
Mortality
Instrumentation
Testing
History
Interactions
Diffusion
No Threat
After viewing the theft and speaking with the experimenter as
well as campus security, all participants were asked to return on
day two to discuss what they saw more formally with the police
detective. Unfortunately, a hurricane hit the next day and closed
down campus for two weeks, so participants were unable to return as
quickly as the researchers would have liked. Still, the interviews
commenced three weeks after the theft, and researchers found that
those in the experimental group did not differ in their memories
regarding the presence of a gun compared to those in the control
condition. That is, the mean score that a gun was present was
similar for the experimental group (M = 65%, SD =
11.4%) and the control group (M = 63%, SD =
13.26%).
Maturation
Regression to the mean
Selection
Mortality
Instrumentation
Testing
History
Interactions
Diffusion
No Threat
To see how participants responded to questions about the
presence of a gun immediately after the theft (on day one), campus
security asked the participant how likely it was that the thief had
a gun immediately after the incident. The experiment then proceeded
as usual, with participants in either the experimental or control
group. The next day (day two), the police detective also asked
about whether the participant saw a gun. Data analysis showed that
participants’ answers about the likelihood that the thief had a gun
became more extreme on follow-up questioning by the police
detective. That is, regardless of whether they were in the
control or experimental condition, participants who reported
that the thief had a gun on day one were convinced he had a gun on
day two. That is, the mean score that a gun was present was similar
for the experimental group (M = 65%, SD = 11.4%)
and the control group (M = 63%, SD = 13.26%).
Maturation
Regression to the mean
Selection
Mortality
Instrumentation
Testing
History
Interactions
Diffusion
No Threat
Thinking that hearing about a gun might be more emotionally
stressful for females then males, the researchers made sure that
all female participants were in the control condition while males
heard the conversation about the presence of a gun. The researchers
found that those in the experimental group did incorporate the gun
misinformation into their memories of the theft more than those in
the control condition. That is, the mean score that a gun was
present was much higher for the experimental groups (M =
67%, SD = 12.4%) than the control group (M = 33%,
SD = 11.16%).
Maturation
Regression to the mean
Selection
Mortality
Instrumentation
Testing
History
Interactions
Diffusion
No Threat
The researchers decided to collect all of the data from the
control group early in the semester and then collect data from the
experimental group later. Although the researcher scripted out
exactly how the confederates were supposed to act for both
conditions and what they were supposed to say, the confederates
were still shy and nervous early in the semester. Fortunately, they
grew much more confident in their performance as the semester
progressed. The researchers found that those in the experimental
group did incorporate the gun misinformation into their memories of
the theft more than those in the control condition. That is, the
mean score that a gun was present was much higher for the
experimental group (M = 67%, SD = 12.4%) than the
control group (M = 33%, SD = 11.16%).
Maturation
Regression to the mean
Selection
Mortality
Instrumentation
Testing
History
Interactions
Diffusion
No Threat
Thinking that hearing about a gun might be more emotionally
stressful for females then males, the researchers made sure that
all female participants were in the control condition while males
heard the conversation about the presence of a gun. After viewing
the theft and speaking with the experimenter as well as campus
security, all participants were asked to return on day two to
discuss what they saw more formally with the police detective.
Unfortunately, a hurricane hit the next day and closed down campus
for two weeks, so participants were unable to return as quickly as
the researchers would have liked. Still, the day two interviews
commenced three weeks after the theft, and researchers found that
those in the experimental group did not differ in their memories
regarding the presence of a gun compared to those in the control
condition. That is, the mean score that a gun was present was
similar for the experimental group (M = 65%, SD =
11.4%) and the control group (M = 63%, SD =
13.26%).
Maturation
Regression to the mean
Selection
Mortality
Instrumentation
Testing
History
Interactions
Diffusion
No Threat
To see just how participants responded to questions about the
presence of a gun, campus security asked the participant how likely
it was that the thief had a gun on day one. The police detective
was similarly supposed to ask about whether the participant saw a
gun, but the detective only had time to talk with half of the
participants on day two. To choose which participants to ask back
for this second interview, the researchers looked only at those who
were really confident that they saw a gun on day one (they answered
75% likelihood or higher). The researchers found that those in the
experimental group did not differ from those in the control group
about the likelihood of a gun. That is, the mean score that a gun
was present was similar for the experimental group (M =
67%, SD = 12.4%) and the control group (M = 63%,
SD = 11.16%), though both were ironically less sure about
the presence of a gun than they were the prior day.
Maturation
Regression to the mean
Selection
Mortality
Instrumentation
Testing
History
Interactions
Diffusion
No Threat
The experiment proceeded as normal, with the experimental group
hearing about the potential presence of a gun as the confederates
left the study and the control group not hearing about the gun.
Unfortunately, hearing about the possible presence of a gun
frightened some of the experimental participants, who did not show
up for the second interview with the interviewer on day two. All of
the control participants returned for the second interview. The
researchers found that those in the experimental group did not
differ from those in the control group about the likelihood of a
gun. That is, the mean score that a gun was present during the
second interview was similar for the experimental group (M
= 67%, SD = 12.4%) and the control group (M =
63%, SD = 11.16%).
Maturation
Regression to the mean
Selection
Mortality
Instrumentation
Testing
History
Interactions
Diffusion
No Threat
The experiment proceeded as normal, with the experimental group
hearing about the potential presence of a gun as the confederates
left the study and the control group not hearing about the gun.
Unfortunately, word spread around campus about a man with a gun who
stole a laptop during a study. Unknown to the study researchers,
many potential participants participated in the research anyway,
and kept an eye out for a gun, knowing that they might be asked
about it. The researchers found that those in the experimental
group did not differ from those in the control group about the
likelihood of a gun. That is, the mean score that a gun was present
during the second interview was similar for the experimental group
(M = 67%, SD = 12.4%) and the control group
(M = 63%, SD = 11.16%).
Maturation
Regression to the mean
Selection
Mortality
Instrumentation
Testing
History
Interactions
Diffusion
No Threat