Deck 5: Introduction to Data Screening and Cleaning

Question 1

If you have a large normally distributed dataset with data missing at random then the best value to insert instead of missing data is the:

A)mode
B)median
C)mean
D)Last Observation Carried Forward

Answer 1

mean

Question 2

Missing data inserted by the method of last Observation Carried Forward is usually used with

A)non-random missing data
B)data which is missing completely at random
C)data from clinical trials
D)data from a correlational study

Answer 2

data from clinical trials

Question 3

A dataset has scores ranging from 30-50, except one outlier who has scored 10.The influence of this outlier could best be dealt with by giving him a score of

A)10
B)22
C)29
D)45

Answer 3

29

Question 4

Questions relate to the following dataset.Stress and anxiety are measured from 1 low) to 7 high).

For the 'stress' variable, are there any scores which might cause us anxiety, if so, which?

A)No there are not
B)Yes there is; it is case number 5
C)Yes there is; it is case number 10
D)Yes there is; it is both case number 5 and 10

Question 5

In SPSS, MVA and EM stand for:

A)Missing Variables Analysis and Estimated Means
B)Missing Values Analysis and Estimated Means
C)Many Variables Analysis and Enhanced Means
D)Many Variables Analysis and Extra Means

Question 6

Questions relate to the following dataset.Stress and anxiety are measured from 1 low) to 7 high).

How would you deal with the unusual score for the stress variable? I would

A)leave the score as it is
B)change the score to 2
C)change the score so that it is missing
D)change the score to 5

Question 7

Questions relate to the following dataset.Stress and anxiety are measured from 1 low) to 7 high).

How would you deal with the missing value for anxiety? I would

A)leave the value as missing
B)insert the mean value for anxiety
C)insert the Last Observation Carried Forward
D)any of the above

Question 8

If researchers delete listwise, this is likely to:

A)result in a decreased sample size for the analysis
B)result in an increased sample size
C)have no effects for the analysis which is carried out
D)none of the above

Question 9

Researchers notice that they have missing data.When they look more carefully, they see that the data was missing because some equipment failed, some people were ill, and in some cases the data was not entered.The data are likely to be:

A)missing completely at random MCAR)
B)missing at random MAR)
C)missing non-randomly
D)none of the above

Question 10

Outliers can be assessed by:

A)the mode
B)the median
C)the mean
D)frequency distribution plots