CPS questions given in class (2 more will occur as options in other sections).

 

1. (5pts) The difficulties in estimating harmful effects of excess ionizing radiation include (MTF):

A)           We have few people exposed to the low doses that are of greatest interest to us

B)           There is not an accepted way of combining the effects of different types of radiation for measuring its health effects

C)           The effects of moderate excess doses (e.g., 3 times the annual average) are slight, which makes it difficult to estimate the effects of low doses

D)           In cases of large exposures, time lags occur between radiation exposure and cancer appearance, so that people exposed may die from other causes before they have time to develop a cancer from the radiation

E)            There are no good non-human animal models for estimating the cancer-causing effects of low levels of (excess) radiation in humans.

F)            A proliferation of electronic devices, such as microwave ovens and cell phones, has increased the U.S. population exposure to ionizing radiation in ways that cannot be easily measured.

 

2. (4pts) You conduct a survey of 200 people attending home football games, for all home games during the season.  The poll questions are the same each game (all multiple choice), and each game you have pollsters gathering the same data in two ways from each participant to verify that they have given you their true response. 

 

Which types of errors can plausibly explain any differences you observe between the distributions of responses obtained at different games (multiple T/F)?

 

 

(A) Sampling       (B)  Bias          (C) RPA        (D) None

 

3-6 (3pts each) Which types of errors are indicated in the following? (one answer each)  Your options are

 

A) Sampling

B) Human & Technical

C) Bias

D) RPA

E) None

 

 

3 (3 pts) You weigh yourself on a scale.  It reads 155#, but you know your weight is 150#.  You get off the scale, and it reads 5#.  What type of error is indicated by the difference between your true weight and the weight indicated by the scale?                                                          (A)   (B)   (C)   (D)   (E)

 

4. (3pts)  You reset the scale to 0 and get on it.  Your exact weight is 150.3789 but the scale reads 150.

                (A)   (B)   (C)   (D)   (E)

5. (3pts)   You purchase 1 lotto ticket each week for a year in which the weekly odds of winning are 1/1000.  Your 7th wins but your other 51 don’t.  What type of “error” explains the discrepancy between the week you won and the weeks you didn’t?     (A)   (B)   (C)   (D)   (E)

 

6 (3pts).  You sample Austin for people’s attitudes about defaulting on student loans.

100 UT students:                           51 favor no penalty

100 Austinites, not UT students:     10 favor no penalty

 

Which type of error plausibly explains the big difference between the two groups? (one only)   (A)   (B)   (C)   (D)  (E)

 

 

7. (4pts)   Which types of errors are lessened (or at least detected) with replication (MTF) ?

(A) Sampling    (B)  Human & Technical     (C)  Bias          (D) RPA

 

8. (3pts) The split sample procedure in drug testing is an example of which property of ideal data (one only):

(A) Blind    (B) Replication    (C) Randomization      (D) Standards

 

 

 

Radiation

 

9.  (4 points).  Which are true? (easy)

(A) High doses are used as models of low doses when calculating the cancer risk from radiation exposure

(B) The use of a single measure of radiation exposure for different types of radiation (rem or rad) is a false model, at least because the different types of radiation have different biological effects.  It is therefore NOT a useful model for measuring the harmful effect of radiation.

(C)  The abstract models underlying our understanding of the cancer risk caused by radiation include (i) different types of cancer, (ii) different types of radiation,  and (iii) different doses.

 

10.  (4 pts).  Which are true? (medium difficulty)

(A) The type of cancer most commonly studied in conjunction with radiation has been leukemia because of its relatively short time-lag between radiation exposure and appearance of the cancer (5 years).

(B) Living in Denver increases your background level of radiation compared to living in Austin because of higher levels of radioactive potassium (K) in your bones.

(C) We do not try to estimate elevated cancer rates from people given medical X-rays on a yearly basis because the effects of this excess radiation are too small to measure.

 

 

 

 

 

 

 

 

 

 

Data: Error

(11-13). For each of the following descriptions, indicate the types of error present (the italicized phrase identifies the error). Mark a type of error only if it is definitely present. Do not assume any more than what is explicitly mentioned in the problem. One answer only for each question.

(A) Rounding, Precision and accuracy

Types of Error:

(B) Sampling

 

(C) Human and technical

 

(D) Bias

 

(E) No error is indicated

11.  (4 pts)   Two forms of an advertisement are tested to decide which is better by giving either form to 200 students and then having them fill out a questionnaire that reveals how well the advertisement worked.  The two forms of the ad are assigned to students randomly, and the students are unaware that there are different forms of the ad.  The questionnaire is multiple choice and scored on a machine, whose performance is checked initially by putting through a form whose answers are all known in advance.  .  Overall, the responses to the two ads are slightly different, but a statistical analysis indicates that these differences are not significant. What type of error is used to explain the difference in responses to the two forms of the advertisement?   (one answer only)

(A)       (B)      (C)       (D)      (E) 

 

12. (4 pts) A racing car’s performance on a track is compared with and without a fuel additive.  In multiple trials with and without the additive, the car’s top speed is consistently 3 miles per hour faster with the additive, but fuel use is also higher (per mile) with the additive.  Thus, it is determined that the additive is advantageous only for short races, because more time is taken refueling the car than is gained by the small increase in top speed.  What type of error in data is indicated by the difference in speed of the car with and without the fuel additive? (one only)

(A)       (B)      (C)       (D)      (E)

 

13. (4 pts) Researchers in the late 1800s and early 1900s attempted to measure the “intelligence” of different human races by filling empty skulls with lead shot to determine brain volume.  They knew the race of each skull at the time they were filling it and apparently (and unconsciously) took more care to settle the shot in some skulls – of Caucasians – than in others. Thus, their measures of brain volumes were smaller for the non-Caucasians (larger for Caucasians) than should have been if the procedure had been conducted the same on all skulls. What type of error is the systematic difference in skull measurement between Caucasian skulls and non-Caucasian skulls? (one only)

(A)       (B)      (C)       (D)      (E)

 

Ideal Data

14 (6 pts) Which of the following points about protocols (written procedures for data) are correct? (MTF)

 

A)           Protocols are models with the goal of describing how data are to be gathered and/or analyzed.

B)            If the protocol is followed, the types of errors likely to be present in data can be understood from reading the protocol

C)            The point made in lecture that protocols have become a substitute for data quality in some government  arenas was illustrated with the example of the FBI DNA lab being reviewed only for whether it was documenting on paper that it followed its protocol, not how good its data were.

D)           An important means of reducing errors in data is to change (improve) a protocol 

E)             If followed to the letter, a written procedure minimizes errors in the data.  That is, most error creeps in because the data are not gathered strictly according to the protocol (CPS question)

F)             Even when the protocol is followed, there are usually many subtle ways in which the written protocol does not describe how the data were collected.  This is one way in which the written procedure is a false model (CPS question)

 

 

 

15-16. Do-it-yourself protocol. You are conducting an external review/test of a DNA lab. Your job is to send two tubes to the lab, with labels. There are several options for the content of and label on a tube. You must decide which contents to send and how to label the tubes so that the features of ideal data requested in the question are present from the lab's perspective. If a tube has a person's name on it, the lab can assume that the tube contents belong to the name of the person on the label. If a tube is labeled with a number, the contents are unknown to the lab but known to you. Your options for tube contents and tube labels are:

option

 tube label

Contents in the tube are from

Blood type

Gender

(A)

Sam Brown

Sam Brown

A

Male

(B)

Holly Berry

Holly Berry

B

Female

(C)

Rachael King

Rachael King

O

Female

(D)

#13

Joel Sachs

AB

Male

(E)

#21

Joel Sachs

AB

Male

(F)

#100

Pam Hines

O

Female

(G)

Anna Himler

Anna Himler

B

Female

(H)

#17

Ulrich Mueller

B

Male

(I)

No combination of tubes can satisfy the protocol

 

In the following questions, choose two letters among options (A)-(H) to describe the two tubes that will be sent to the lab. The tube labels are the only information the lab receives about the samples.  If it is possible to satisfy the protocol, the question will require exactly two letters and only two letters -- one for each tube. Thus, the answer for a question might be (A) & (B), or it might be (D) & (F). If more than one pair of options are possible correct answers, fill in only one correct pair of options. Thus, if (A) & (B) is one acceptable answer, and (C) & (D) is another acceptable answer, fill in either (A)&(B) or (C)&(D), but not both.   If a factor (such as identity, blood type, gender) is not specified in the protocol, then that factor will be ignored in grading the answer.  Alternatively, if a protocol cannot be satisfied with two from (A)-(H), fill in (I).

 

15. (3 pts) Choose two tubes to guarantee replication of blood type but not individual or gender; the replication of blood type should be fully blind to the lab – that is, the lab should not be able to tell from the information on the tubes that the two samples have the same blood type.

two tubes:                             (A)          (B)           (C)           (D)          (E)           (F)           (G)           (H)          (I)

 

16. (3 pts) Make the tube identities fully blind to the lab and replicated for gender and blood type but not replicated for individual.

two tubes:                             (A)          (B)           (C)           (D)          (E)           (F)           (G)           (H)          (I)

 

 

 

17.(4 pts). A recent change to the class protocol for collecting and recording homework has been implemented this semester because (one only)

A)    The instructors found a new way to streamline the process

B)     Too many homework grades were being challenged

C)    An error occurred in keeping track of all homeworks.

D)    No change in class protocol has been implemented this semester.

 

 

(18, 19). For each of the following statements, mark the appropriate letters that describe the data design features present. Mark a data feature only if it is explicitly present in the problem description. MTF

 

(A) explicit protocol

(C) standards

(E) blind

(B) replicates

(D) random

(F) none

 

18. (4pts) Two forms of an advertisement are tested to decide which is better by giving either form to 200 students and then having them fill out a questionnaire that reveals how well the advertisement worked.  The two forms of the ad are assigned to students randomly, and the students are unaware that there are different forms of the ad.  The questionnaire is multiple choice and scored on a machine, whose performance is checked initially by putting through a form whose answers are all known in advance.  Overall, the responses to the two ads are slightly different, but a statistical analysis indicates that these differences are not significant.  What features of ideal data are indicated?  (yes, this is the same as one used above for “error”)

(A)          (B)           (C)           (D)          (E)           (F)

 

19. (4 pts)  A police group decides to determine how gullible the public is to scams.  Hundreds of people are sent invitations to seminars about how to obtain government subsidies (“free money”) for home improvement; of course, no mention is made in this invitation about the real purpose.  Each person is invited to only one seminar, but different people are invited to different seminars, so that the police group can try different scamming methods and see which are the most effective.  Approximately 50% of the invitees attend their seminar, where they are encouraged to provide the seminar organizers with social security numbers, bank account numbers, and other data that a con artist could use.  The attendees are then informed of their vulnerability to scams.   Which features are indicated?

(A)          (B)           (C)           (D)          (E)           (F)

 

Drug Testing

               

20. ( 5 pts). What constitutes a standard in a drug test for evaluating lab error rates?  (MTF)

A)     A sample with a known level of drug present.

B)     A sample known to be drug-free.

C)     A written procedure describing the level of performance to be upheld by the lab

D)     Any measure taken by the lab to detect or reduce human and technical error

E)      A proficiency test given to the lab that does the analysis, regardless of whether the test is blind.

 

21 (5 pts) Which of the following properties of ideal data are mandated by for DoT drug-testing procedures? MTF

A)     Standards are not part of the protocol but are often tested anyway

B)     The test procedure must be replicated for all positive samples

C)     The samples must be coded so that their identities are unknown to the lab

D)     A person who tests positive on a first test may request a second test

E)      Samples are to be tested in a random order

F)      Monthly proficiency tests are conducted by an outside agency

 

DNA Typing

22 (4pts) Which of the following properties apply to DNA and/or DNA typing? (MTF)

A)    The RMP (random match probability) is typically less than 1/billion for the best methods

B)     In the Castro case, several lab errors and violations of protocol were discovered that contributed to a gross miscalculation of the RMP.

C)     DNA typing using mitochondria usually has larger RMPs (closer to 1.0) but can use poorer DNA samples than STR methods.

D)    Blood is a poor source of DNA for typing because our red blood cells do not have nuclei.

 

23 (5pts) Changes to DNA typing methods since its inception include which of the following (MTF):

 

A) The RMP calculated in the famous, first DNA typing case in Britain was only about 1/1000; it is now much, much smaller (1/billion or better).

B) The basic principle underling DNA typing has not changed – that we all have differences in our DNA that make us unique (except for identical twins).

C) The DNA barcode is now read by a machine, so false matches in DNA typing cases have almost been eliminated.

D) Most labs now subject themselves to blind proficiency testing

E) The Houston Crime lab was recently exposed for major problems in its DNA analyses.

 

24-26 (3 pts each). Fill in the blanks of the “consequences” column of the table with the best single option (most specific) from the list below the table.  The question number is given in each blank.  One answer only per question.

 

Deviation from Ideal Data

Consequences

Samples not handled blindly

24.

Insufficient replication of the DNA typing process in a case

25.

Inadequate protocols for analysis of results

26.

Your choices for consequences are:

(A) Improper calculation of RMP (random match probability) in some cases

(B) DNA sample degradation occurs (loss of DNA quality)

(C) The RMP threshold for conviction will appear to be exceeded when it is actually not exceeded

(D) Sample mix-ups in a case can go undetected

(E) This protocol is no longer applicable because of recent changes in DNA typing methods

(F) The protocol allows for a biased willingness to accept results

(G) The STR (short tandem repeat) RMP will not match the mitochondrial RMP

(H) The protocol increases the likelihood of sample mix-up

(I) The full extent of lab error rates remains unknown

(J)  This protocol allows outliers of natural variation to escape detection

 

27. (4 pts.) Exam Key Code: Fill in (AB) on question 27 to indicate your exam code.  Also, fill in the correct bubbles for your name and pad number on the scantron form.