Morganparkcps.enschool.org

AP® STATISTICS
2011 SCORING GUIDELINES (Form B)
Question 2

Intent of Question

The primary goals of this question were to assess students' ability to (1) distinguish an experiment from an
observational study; (2) critique statistical information, in particular whether or not researchers are justified
in making a specific conclusion based on the given information; (3) recognize and describe a potential
problem with a study that lacks random assignment or blinding.
Solution

Part (a):

The study was an experiment because treatments (D-cycloserine or placebo) were imposed by the researchers on the people with acrophobia.
Part (b):

No, the experiment was designed to compare the D-cycloserine group with a control group that received the placebo. The researchers can conclude that the D-cycloserine pill and two therapy sessions show significantly more improvement than a placebo and two therapy sessions. However, there is no basis for comparison with another group of people with acrophobia who received eight therapy sessions and no pill.
Part (c):

One example is that if the therapists were allowed to choose who received the placebo and who received D-cycloserine, they might assign the people with more severe acrophobia to one of the groups and the people with less severe acrophobia to the other group. Thus, the improvement after only two therapy sessions could be related to the initial severity of the acrophobia rather than to the effects of D-cycloserine.
Scoring

Parts (a) and (c) are scored as essentially correct (E), partially correct (P), or incorrect (I). Part (b) is scored as
essentially correct (E) or incorrect (I).
Part (a) is scored as follows:
Essentially correct (E) if the response indicates that this was an experiment, AND the explanation clearly communicates that two treatments were imposed. Partially correct (P) if the response indicates that this was an experiment, BUT the explanation does not clearly communicate that two treatments were imposed. Note: If the response indicates that this was an experiment because there was random assignment to treatments, this implies imposition of treatments and is scored as E. If the response does not clearly state that the random assignment is to treatments, this is scored as P. Incorrect (I) if the response indicates that this is an observational study OR if the explanation is missing or incorrect. 2011 The College Board. Visit the College Board on the Web: www.collegeboard.org. AP® STATISTICS
2011 SCORING GUIDELINES (Form B)
Question 2 (continued)

Part (b) is scored as follows:
Essentially correct (E) if the response says "no" AND clearly explains why this is not reasonable based on the fact that there was no experimental group that received eight therapy sessions and no pill. Incorrect (I) if the response provides an answer with an incorrect or no justification.
Part (c) is scored as follows:
Essentially correct (E) if the response indicates that this method of assignment might create experimental groups that differ in some systematic way other than the treatment AND provides a justification that describes the potential confounding, if the response indicates that if the therapists know who was in which group, it may influence the therapists' behavior when dealing with or evaluating the people with acrophobia. Partially correct (P) for any of the following: • The response indicates that the assignment might create experimental groups that differ in some systematic way, BUT does not provide an explanation of the potential confounding. • The response makes a general statement that the lack of random assignment could lead to confounding BUT does not provide an example in context. • The response indicates that the therapists know who is in which group BUT does not give a reason as to why this might lead to a misleading conclusion. • The response makes a general statement that failure to blind may lead to bias BUT does not provide an example in context. Incorrect (I) if the response fails to meet the criteria for E or P. Note: If the response discusses incorrect conclusions that might result from having the people with acrophobia (rather than the therapists) choose their own treatments, the response is scored as I, because such a response does not address the question asked. Complete
Response
All three parts essentially correct Substantial
Response
Two parts essentially correct and one part partially correct Developing
Response
Two parts essentially correct and one part incorrect One part essentially correct and one or two parts partially correct. 2011 The College Board. Visit the College Board on the Web: www.collegeboard.org. AP® STATISTICS
2011 SCORING GUIDELINES (Form B)
Question 2 (continued)
Minimal Response
One part essentially correct and two parts incorrect One or two parts partially correct and one part incorrect 2011 The College Board. Visit the College Board on the Web: www.collegeboard.org. 2011 The College Board.
Visit the College Board on the Web: www.collegeboard.org.
2011 The College Board.
Visit the College Board on the Web: www.collegeboard.org.
2011 The College Board.
Visit the College Board on the Web: www.collegeboard.org.
AP® STATISTICS
2011 SCORING COMMENTARY (Form B)
Question 2

Sample: 2A
Score: 4

In part (a) the response indicates that subjects were assigned treatments by the researchers and that there
was more than one treatment. Part (a) was scored as essentially correct. Part (b) is answered concisely and
correctly by noting that the eight therapy sessions without the pill was not one of the treatments included
in the study. Part (b) was scored as essentially correct. The first sentence in part (c) correctly indicates that
allowing the therapists to assign people to groups might result in experimental groups that differ in "fear
intensity," but it is not accompanied by an explanation of how this potential confounding could lead to an
incorrect conclusion. However, the second and third sentences in part (c) indicate that allowing the
therapists to assign people to groups would compromise double-blinding, with the consequence that the
therapists might "adjust the therapy accordingly." The statement that the therapists' behavior when
dealing with the people with acrophobia might be influenced by their knowledge of treatment
assignments was considered adequate explanation of why an incorrect conclusion might be reached, and
part (c) was scored as essentially correct. Because three parts were scored as essentially correct, the
response earned a score of 4.

Sample: 2B
Score: 3

In part (a) the response indicates correctly that the study was an experiment because the independent
variable was "manipulated" and the dependent variable was "to be observed." The response also indicates
that there was a "control group and experimental group." Part (a) was scored as essentially correct. Part (b)
was answered correctly by noting that because the study includes only treatments with two therapy
sessions, conclusions can be made only about treatments with two therapy sessions. Part (b) was scored
as essentially correct. The response in part (c) indicates that lack of random assignment could lead to
confounding, but no example in the context of the study is provided. Part (c) was scored as partially
correct. Because two parts were scored as essentially correct and one part was scored as partially correct,
the response earned a score of 3.
Sample: 2C
Score: 2
The response in part (a) indicates that the study was an experiment "because the researcher directly
introduced a variable (the drug) to see its effects on the subjects." Although this response suggests
imposition of a treatment, it does not clearly communicate that two treatments were imposed, and
part (a) was scored as partially correct. In part (b) the response incorrectly states that the researchers would
be justified in reaching a conclusion about eight therapy sessions, so part (b) was scored as incorrect. The
response in part (c) indicates that if therapists are permitted to assign the people with acrophobia to
treatment groups, they might assign those with less fear to the D-cycloserine group, which could result in
overestimation of the effectiveness of the drug. This is a clear explanation of potential confounding, and
part (c) was scored as essentially correct. Because one part was scored as essentially correct, one part was
scored as partially correct, and one part was scored as incorrect, the response earned a score of 2.
2011 The College Board. Visit the College Board on the Web: www.collegeboard.org.

Source: http://morganparkcps.enschool.org/ourpages/auto/2014/3/17/52515000/ap11_statistics_form_b_q2.pdf