 |
1 | |
A market researcher performs the following tasks while conducting a study. Which of the tasks doesn't belong in the data preparation process? |
|
| | A) | She makes sure that the data collected is valid and free of bias. |
| | B) | She conducts preliminary data analysis to do an informal check if the hypotheses are supported by data or not. |
| | C) | She edits the data and codes the responses from the survey responses. |
| | D) | She enters the data into an Excel file from all the paper surveys. |
| | E) | She prepares data tables to check how many responses fall in different categories. |
|
2 | |
Curbstoning is: |
|
| | A) | a data validation technique where a supervisor randomly pick completed surveys and re-contacts the respondents to figure out if their responses have been correctly noted. |
| | B) | a data collection technique similar to "mall intercept" where respondents are approached at mall curbs and asked if they would be willing to participate in the survey. |
| | C) | when interviewers fill out the surveys themselves rather than with an actual respondent. |
| | D) | a data validation technique where a researcher makes sure that data encoding is in accordance with the suggested techniques. |
| | E) | None of the above. |
|
3 | |
A part of the data validation process is to make sure that the data were collected from qualified respondents only. This area of data validation is referred to as: |
|
| | A) | Screening |
| | B) | Curbstoning |
| | C) | Procedure |
| | D) | Sampling frame |
| | E) | Population frame |
|
4 | |
Surveyors responsible for data collection for a study are told that they must ask respondents questions about their purchases right after they leave a grocery store. In order to make sure that the data were collected under these exact circumstances, a researcher is likely to do a validation check for: |
|
| | A) | Fraud |
| | B) | Procedure |
| | C) | Courtesy |
| | D) | Completeness |
| | E) | Screening |
|
5 | |
Which of the following is not included in the data editing step? |
|
| | A) | Make sure that the proper questions were asked of the respondents. |
| | B) | Make sure that the responses were recoded accurately. |
| | C) | Make sure that all the responses were entered correctly in the database. |
| | D) | Make sure that (based on responses to screening questions) data were collected from only those who were eligible to participate in the survey. |
| | E) | Make sure that the responses to open-ended questions were interpreted correctly. |
|
6 | |
Which of the following statements (regarding the sequencing of steps for developing codes for responses for open-ended questions) is true? |
|
| | A) | The process begins by assigning a code value for each response (e.g., each questionnaire is assigned a numerical code). |
| | B) | After giving numerical code to each questionnaire, the researcher should give code numbers to each response category. |
| | C) | After generating a list of all possible responses, responses are consolidated into categories that convey the same shared meaning. |
| | D) | Numerical codes are assigned to each response category as the last step of the process. |
| | E) | All of the above statements represent correct ordering. |
|
7 | |
There were 22 different responses given for an open-ended question. However, when the coding was done for this question, there were only 6 separate response codes assigned for the question. This consolidation of responses must have occurred during which of the following steps of the code development process: |
|
| | A) | Step 1 |
| | B) | Step 2 |
| | C) | Step 3 |
| | D) | Step 4 |
| | E) | Step 5 |
|
8 | |
All of the following approaches can help in error detection for data entry, EXCEPT: |
|
| | A) | numerical variable analysis |
| | B) | visual verification of the entered data on a computer screen |
| | C) | error edit routines |
| | D) | visual verification of printed representation of the entered data |
| | E) | visual verification of data/column list for the entered data |
|
9 | |
The rows in a data/column list (for data entry error detection) represent: |
|
| | A) | the number of respondents included in the dataset |
| | B) | the frequency of responses for each data field |
| | C) | the number of respondents for which there is error-free data available in the dataset |
| | D) | the number of data fields included in the dataset |
| | E) | None of the above |
|
10 | |
One-way tabulations can be used for all of the following, EXCEPT: |
|
| | A) | They can be used to determine the amount of non-response to individual questions. |
| | B) | They can be used to determine if certain variables differ when compared among various subgroups of the total sample. |
| | C) | They can help locate major data entry errors. |
| | D) | They can help determine descriptive statistics such as means and standard deviations. |
| | E) | They can be used to communicate the results of the research projects. |
|
11 | |
All of the following are true about cross-tabulations, EXCEPT: |
|
| | A) | They are typically done after doing the one-way tabulations. |
| | B) | They are good for studying relationship between variables. |
| | C) | They show percentage of responses for both rows and columns |
| | D) | Demographic or psychographic variables are typically the starting points in developing cross-tabulations. |
| | E) | They are good for calculating standard deviations of two or more variables simultaneously. |
|
12 | |
Once data has been entered into a spreadsheet, the next task in the data preparation stage is data editing and coding. |
|
| | A) | True |
| | B) | False |
|
13 | |
A research supervisor calls back 10% of a survey's respondents to check if they actually participated in the survey (to make sure that the surveyor had not falsified any data). This callback is done to counter curbstoning. |
|
| | A) | True |
| | B) | False |
|
14 | |
Data tabulation can be used for the purpose of validation of the accuracy of the data. |
|
| | A) | True |
| | B) | False |
|
15 | |
A researcher wants to determine if there is missing data for one variable for a respondent, will there be missing data for another variable from the same respondent. A way to determine this is to use one-way tabulations. |
|
| | A) | True |
| | B) | False |