Deck 16: Data Preparation for Analysis

Question 1

Which of the following is FALSE with respect to the coding of open-ended questions?

A)The use of several coders can lead to inconsistent treatment of answers.
B)Open-ended questions are generally more difficult to code than closed-ended questions.
C)The coder must determine categories on the basis of answers that are not always anticipated.
D)Coding open-ended questions is typically less expensive than coding closed-ended questions.
E)When the task requires multiple coders,each coder should be assigned parts of the questionnaire for all questionnaires rather than a subset of the questionnaires.

Answer 1

D

Question 2

During the editing process,the analyst must

A)ensure units used for open-ended question are consistent.
B)decide how to deal with cases that have incomplete answers.
C)determine what to do about cases with obviously wrong answers.
D)All of these are correct.
E)None of these are correct.

Answer 2

D

Question 3

Which of the following statements is(are)TRUE regarding coding?

A)The classes should always be mutually exclusive and exhaustive.
B)Multiple responses should never be coded.
C)Coding closed-ended questions is more difficult than coding open-ended questions.
D)Alphabetic codes should be assigned to the classes.
E)All of these are true statements.

Answer 3

A

Question 4

The editing process involves

A)the inspection of the data received from each sample element.
B)the correction of the data received from each sample element.
C)the transformation of the raw data received from each sample element into symbols.
D)All of these are correct.
E)the inspection and correction of the data received from each sample element.

Question 5

Which of the following might NOT be an appropriate strategy for dealing with missing information on a questionnaire?

A)Throw out the entire questionnaire
B)Overlook the missing information and code the remaining answers
C)Substitute information based on the responses of similar respondents
D)None of these are appropriate.
E)All of these might be appropriate.

Question 6

Which of the following are valid software applications an analyst might use to build the data file?

A)Word processing software
B)Database software
C)Spreadsheet software
D)Statistical packages such as SPSS
E)All of these are correct.

Question 7

If a respondent indicates that he drives a foreign car,then later,in the same questionnaire,identifies it as a Ford,the editor should

A)change the "Ford" identification to "unknown foreign".
B)change the "foreign" identification to "domestic".
C)throw out both responses.
D)determine which of the responses is correct.
E)Any of these may be correct.

Question 8

The main aim of editing is to

A)ensure that the analysis is valid.
B)establish minimum quality standards for the raw data.
C)establish a balance between costs and accuracy.
D)establish codes for the raw data.
E)impose maximum quality standards on the raw data.

Question 9

The most difficult type of responses to code are

A)close-ended,multiple response (e.g.,"Check all that apply")questions.
B)open-ended,exploratory questions.
C)close-ended,single response questions.
D)open-ended,factual questions.
E)None of these are correct.

Question 10

The following categories of ages are ____ and ____,but not ____: "18-24/25-34/35-44/45-54/55 and over"

A)closed-ended,exhaustive,mutually exhaustive
B)open-ended,mutually exclusive,exhaustive
C)closed-ended,mutually exclusive,exhaustive
D)exhaustive,mutually exclusive,open-ended
E)None of these are correct.

Question 11

You are coding a survey that asks the question "Please tell us your age: _____",where the respondent fills in the blank with a number corresponding to his or her age. The best way to code this information is

A)using narrow categories,such as the following: Under 18,19-21,22-25,26-29,30-33,etc.
B)recording the actual age.
C)using broad categories,such as the following: Under 20,20-29,30-39,40-49,etc.
D)None of these are correct.
E)All of these are correct.

Question 12

When coding data for computer analysis,which of the following is NOT recommended?

A)Use only one symbol per column in the computer record
B)Use alphabetic characters
C)Assign as many columns to a variable as is necessary to capture the variable, e.g., two columns would need to be assigned to a question with 20 possible answers
D)Use standard codes like "8" for all "No answers" and "9" for all "Don't know" throughout the study
E)Leave data ungrouped

Question 13

Which of the following questions do you think would be the easiest to code?

A)What are the three characteristics that you find most pleasing when using product X?
B)Have you ever used product X? ____ Yes ____ No
C)What religious denomination do you consider yourself?
D)Please specify the type of television set in your home?
E)How do you feel about commercials on children's TV shows?

Question 14

You have just completed data collection for a research project.Which of the following is the first thing you should do with the data?

A)Build a data file
B)Determine the categories or classes you'll use to code open-ended questions
C)Examine the data to detect and resolve incorrect,missing,or incomplete responses
D)Run frequencies on the data to check for blunders
E)Develop your codebook to document how raw data is coded in the data file

Question 15

Which of the following statements about the editing process are TRUE?

A)When it's obvious a respondent hasn't taken the study seriously,his or her answers should be dropped.
B)All completed questionnaires should be kept.
C)All incomplete questionnaires should be dropped.
D)You can't change incorrect answers because you have no way of knowing the respondent's original intent.
E)All partially complete questionnaires should be kept.

Question 16

Which of the following is NOT a recommended coding convention?

A)Use as many columns as necessary for the field
B)Locate only one character in each column
C)Use alphabetic codes if possible
D)Use consistent codes for similar types of responses
E)Code in an identification number for each questionnaire

Question 17

Coding transforms raw data into ____ that may be ____.

A)secondary data,tabulated
B)symbols,manipulated
C)accessible form,automatically retrieved
D)tables,easily perceived
E)tables,analyzed for relationships between variables

Question 18

The location of each variable in the data array and the way in which it was coded are contained in a

A)diary.
B)random file.
C)codebook.
D)focus group.
E)catalog.

Question 19

Which of the following is NOT one of the steps that must happen before data can be analyzed?

A)Editing the data
B)Coding the data
C)Interpreting the data
D)Building the data file
E)Cleaning the data

Question 20

The purpose of the coding process is to

A)transform raw data into symbols.
B)encrypt the raw data so that it is secure from unauthorized use.
C)determine if the raw data meets minimum quality standards.
D)detect incorrect or invalid responses.
E)separate completed questionnaires from incomplete ones.

Question 21

Which of the following strategies for handling missing data makes maximum use of the data?

A)Substituting values for the missing data
B)Reporting the number of blanks as a separate category
C)Eliminating the case with the missing data in analyses using the variable(s)for which data is missing
D)Eliminating questionnaires with missing data
E)None of these are correct.

Question 22

A respondent indicated she redeemed a coupon at Walmart last week but later indicated that she had not visited a Walmart in over two weeks.This type of response poses a problem of

A)completeness.
B)legibility.
C)comprehensibility.
D)consistency.
E)uniformity.

Question 23

In a multiple-column record of a data file,____ represent different variables and ____ represent different respondents.

A)codes,numbers
B)numbers,codes
C)rows,columns
D)columns,rows
E)codes,symbols

Question 24

In a job satisfaction survey,respondents were asked to indicate how many years they had worked for the company.One respondent wrote "ten months" in response to the item.A conversion of the answer to the correct unit of time (years)would be most likely to take place during the

A)building of the data file.
B)editing process.
C)coding process.
D)data analysis.
E)collection phase.

Question 25

At a minimum,a codebook should include all of the following EXCEPT

A)the results of the study.
B)the variable name to be used in statistical analyses for each variable included in the data file.
C)the column(s)in which each variable is located in the data file.
D)a description of how each variable is coded.
E)an explanation of how missing data are treated in the data file.

Question 26

Tom was given the task of assigning numbers to each of the classification categories.For example,Freshman were assigned a 1,Sophomore a 2,and so on.This is known as the process of assigning.

Question 27

After supervising an entire study,John went through every questionnaire to check for completeness,legibility,comprehensibility,consistency,and uniformity.John was conducting what is known as editing.

Question 28

Which of the following statements about the coding process is FALSE?

A)During the coding process,data is categorized.
B)Raw data are transformed into symbols during the coding process.
C)Coding involves judgment on the part of the coder.
D)The coding process occurs almost automatically.
E)All of these statements about the coding process are true.

Question 29

What is the best way to code data from a survey that contains many open-ended,exploratory questions in order to reduce bias?

A)Use the most experienced researcher to do the coding
B)Use two researchers to do the coding and have them compare their results
C)Use any researcher but have him or her double-check the coding
D)Use computer software to do the coding
E)Use an outside researcher who is unfamiliar with the project to do the coding

Question 30

Treatment of missing data depends mainly on the

A)purpose of the study.
B)incidence of the missing items.
C)methods that will be used to analyze the data.
D)None of these are correct.
E)All of these are correct.

Question 31

Which of the following statements about open-ended questions is FALSE?

A)Precoding is not necessary.
B)Response categories are provided for respondents.
C)There are multiple legitimate responses.
D)When categorizing open-ended responses,it is often necessary to include an "other" category.
E)All of these statements about open-ended questions are true.

Question 32

Which of the following is NOT a step in coding open-ended responses?

A)Identify separate responses given by each individual
B)Specify categories into which the responses can be placed
C)Place each response into as many categories as possible
D)Assess the degree of agreement between multiple coders
E)All of these are steps in the process of coding open-ended responses.

Question 33

Which of the following is NOT a recommended practice for coding data and entering it into a file?

A)Assign specific column locations for particular variables
B)When a question allows multiple responses,use the same variable for each response option
C)Use only numeric codes,not letters of the alphabet or special characters like @
D)Use standard codes for "no information"
E)Code a respondent identification number on each record

Question 34

Multiple coders are recommended for all of the following reasons EXCEPT:

A)The use of multiple coders can shorten the amount of time it takes to code.
B)Multiple coders can help reduce bias in the interpretation of different responses.
C)Results can easily be compared to ensure agreement among coders.
D)The use of multiple coders makes the coding process less expensive.
E)All of these are legitimate reasons for using multiple coders.

Question 35

In descriptive research,most of the items included in a questionnaire are likely to be

A)precoded.
B)closed-ended.
C)open-ended.
D)exhaustive.
E)mutually exclusive.

Question 36

A data-entry operator was having a bad day while inputting data from your research project.He occasionally entered "9" when meaning to enter "3".This is an example of a(n)

A)excusable error.
B)blunder.
C)codebook error.
D)outlier.
E)nonresponse error.

Question 37

A questionnaire uses a 1-5 Likert scale to determine job satisfaction.When entering the data into a file,a researcher types a 7 instead of the 4 that the respondent had circled on the questionnaire.Such a mistake can be uncovered by performing what type of analysis?

A)Analysis of variance
B)Regression and double-entry
C)Double-entry
D)Frequency analysis
E)Double-entry and frequency analysis

Question 38

Which of the following might indicate that a respondent lacked interest in a questionnaire?

A)Check marks are not within the boxes provided.
B)Scribbles on the questionnaire
C)Spills on the questionnaire
D)It's unlikely that these mistakes occur due to lack of interest.
E)All of these examples may indicate a lack of respondent interest.

Question 39

The BEST way to handle missing items when analyzing the data is to

A)leave the item blank and report the number blank as a separate category.
B)eliminate the case with the missing item in analyses using the variable.
C)substitute values for the missing item.
D)eliminate the case from all further analyses.
E)There is no single best way for handling missing items.

Question 40

Which of the following is a recommended strategy for handling missing data?

A)Eliminate case(s)with missing item(s)from all further analyses
B)Eliminate the case with the missing item in analyses using the variable
C)Substitute values for the missing items
D)Contact the respondent again
E)All of these could be used.

Question 41

Compare and contrast the various methods or options for dealing with missing data in analyses.

Question 42

The use of scanner technology to "read" responses on paper surveys and to store these responses in a data file is called optical scanning.

Question 43

The coding process involves considerable effort on the part of the coder when the question type is closed-ended.

Question 44

The most difficult questions to code are questions using the Likert scales.

Question 45

In descriptive research,most of the items included in a questionnaire are likely to be open-ended.

Question 46

Coding is the process of transforming raw data into symbols.

Question 47

Optical scanning uses scanner technology to "read" responses on paper surveys and then stores these responses in a data file.

Question 48

On most projects,frequencies should initially be run on all variables to help identify blunders.

Question 49

Item nonresponse is a source of nonsampling error that arises when a respondent agrees to an interview but refuses,or is unable,to answer specific questions.

Question 50

Blunders are errors that occur during editing,coding,and especially data entry.

Question 51

While not a formal rule,if half or more of the responses are missing on a survey,it is recommended to drop that case entirely.

Question 52

An error that arises during editing,coding,or data entry is called a blunder.

Question 53

The codebook is essentially a map to help the researcher navigate from data collection to data editing.