Deck 6: The Beast of Bias

Question 1

Which of the following is not an example of a probability sample?

A)Asking listeners to call in and vote in a football radio phone-in poll with either a yes or no response.
B)Selecting every 5th name on a list of 100 participants in alphabetical order in order to randomly sample 20% of the group.
C)A random sample of hospital patients is selected to receive a questionnaire about the helpfulness of volunteers on the children's ward.
D)A random sample of participants by age group (20-24,25-29,30-34,35-39)selected as a ratio of the overall population.

Answer 1

A

Question 2

In studying the national concept of the word 'character' in relation to morals and virtues,a researcher picks a simple random sample of 100 local flats and visits them to interview the occupants.After the first follow-up visit,the researcher has not been able to get in touch with the occupants in 19 of the properties.The researcher then chooses a second sample of 100 flats in a different area and continues until 100 responses have been received.What assumption affects any results which can be drawn from this study?

A)Human error
B)Overestimate of average results
C)Undercoverage bias
D)Reader-response bias

Answer 2

C

Question 3

Sampling error can only be eliminated when conducting a questionnaire of students and every question is a closed question.

Answer 3

False

Question 4

In the advice column of a national student magazine,final-year undergraduate readers are asked: 'If you could do it again,would you still go to university?' Of the more than 1000 responses,55% said no.What does this show?

A)The survey is meaningless because of voluntary response bias.
B)Nothing can be inferred as nothing is known of the personalities of the respondents.
C)To improve the significance of the study,a random sample of respondents should have been carried out.
D)The survey would be more significant is respondents were called to a focus group.

Question 5

To survey the opinions of football fans at Newcastle United's St.James' Park ground,you and your team of researchers plan to select every hundredth fan entering the stadium via the East Stand turnstiles at the next home match.Is this a simple random sample?

A)Yes,because each fan has the same chance of being selected.
B)Yes,because the 99 out of 100 fans who are not selected will form a control group.
C)Yes,this is an elaborate form of simple random sample.
D)No,because not every sample of the intended size has an equal chance of being selected.

Question 6

Use of a large number of subjects to control for small-sample variability is of paramount importance when designing a study.

Question 7

A researcher is asked to contact 125 members of academic staff in a university at random using the university telephone directory.This is achieved by opening the pages at random and selecting the first 125 names on the page.This is an example of?

A)Chance sampling
B)A simple random sample
C)Selection bias
D)Strategic sampling

Question 8

Sampling error concerns natural variation between samples,and is always present.

Question 9

There is no such thing as a 'bad sample'.

Question 10

Another researcher is planning a different survey of families,this time,all from Northumberland.Using census lists and details of the major towns in Northumberland,the sampling procedure is to obtain a random sample of families from each of the major towns and cities in the county.This type of study will give comparative information which a simple random study cannot?

Question 11

Careful analysis of a sample will indicate whether or not it was randomly selected.

Question 12

Sampling techniques which use probability techniques are successful ways to eliminate participant bias.

Question 13

Data obtained from a sampled group of willing participants are always more accurate than data obtained from existing census information,regardless of the construction of the study?

Question 14

Data were collected in 15 villages on the percentage of women housewives in 2008 and again in 2012.The study was concerned whether this percentage had either increased or decreased.What kind of design was this?

A)A matched-pairs study
B)An observational study
C)A matched-pairs,observational study
D)An experiment using a block design

Question 15

Sampling error suggests an error has been created by the data inputter?

Question 16

Each of the 20 English Premier League teams have a match day squad of 18 players.A sample of 40 players is to be chosen.Each team is asked to put the 18 players' names into a bag and randomly draw out two names.The two names from each team are then combined to make up the sample.Does this method result in a simple random sample of the 360 footballers?

A)Yes,because each player has the same chance of being selected.
B)Yes,because each team is equally represented.
C)Yes,because there are squads of players used,not just the first choice 11.
D)No,because not every group of 40 players has the same chance of being selected.

Question 17

Which of the following is true of voluntary response samples?

A)Voluntary response samples often underrepresent people with strong opinions.
B)Voluntary response samples are only possible with postal surveys.
C)Voluntary response samples have no opt in or out choice.
D)Voluntary response samples are 100% reliable.

Question 18

Your head of department asks you to administer a survey to undergraduates about teenage pregnancies.You believe that male and female students might react differently to the survey questions as they are deliberately provocative.The male and female students are randomly assigned in turn to two separate groups.One of the two groups is asked the provocative questions and the other group are asked less provocative ones as a 'placebo'.Based on the above information,what has happened to the design of the study?

A)It is completely randomized.
B)It is blocked by gender.
C)It is completely randomized,and blocked by gender.
D)It is a matched-pairs design.

Question 19

A double-blind design is important in an experiment because there is a natural tendency for subjects in an experiment to want to please the researcher.

Question 20

You are asked to survey undergraduate students in the education department of your institution.You decide select a simple random sample.You get permission to use 200 participants.Which of the following would be the best approach to select your sample?

A)Print the names of every eligible student,place them in a pot and draw the first 200 out at random.
B)Select two lectures at random and include the students who are in attendance
C)Select from a list one name at random,then select every 10th student after that until 200 are selected.
D)Select the first 200 students to visit your office when you send a message advertising free cakes.

Question 21

In a survey about autism in children,a participant doesn't understand one of the questions and,instead of seeking clarity,answers at random.This is an example of?

A)Sampling error
B)Human error
C)Response bias
D)None of the above

Question 22

You design a questionnaire to survey people's opinions about the increased tuition fees for undergraduate students at UK universities and administer it to 100 new undergraduate students,all of whom have paid the increased fee in 2012.You receive a 100% response rate.This is an example of?

A)A well-designed survey
B)Chance
C)A poorly designed survey because of selection bias in the sampling
D)None of the above

Question 23

The undergraduate population of the University of Birmingham is formed of 25% ethnic minorities (non-white).When sampling 20 students at random,5 are from ethnic minorities.This shows?

A)Sampling error
B)Human error
C)Response bias
D)None of the above

Question 24

A research administrator accidently records data obtained from the above survey incorrectly.This is an example of?

A)Sampling error
B)Human error
C)Response bias
D)None of the above