Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
50 Cards in this Set
- Front
- Back
2 types of errors in survey research |
Sampling (controlled)
Nonsampling |
|
Nonsampling error |
All errors in a survey except those attributable to the sample plan and size.
Cannot be measured by a formula. |
|
Nonsampling error includes: (5) |
1. All types of nonresponse error 2. Data gathering error 3. Data handling errror 4. Data analysis error 5. Interpretation error
(prob definition and question wording)
(mistakes are common) |
|
Data collection |
When respondents provide answers/info to inquiries posted to them by the researcher. Can increase the amount of nonsampling error but minimize the effects. |
|
2 types of errors in data collection |
Fieldwork
Respondent
|
|
Fieldwork error |
Errors committed by the individual who administer questionnaires/interviews.
(can be dependent on resources)
-Intentional and unintentional errors |
|
Respondent error |
Errors on part of the respondent.
(Some methods of data collection have more error than others)
-Intentional and unintentional errors |
|
Intentional fieldwork errors (2 types) |
Collection person willfully violates the data collection requirements set fourth by the researcher.
- Interviewer cheating and leading respondent |
|
Interviewer cheating |
Interviewer intentionally misrepresents respondents.
-Turn in uncompleted interviews as complete or interviews convenient people instead of designated person.
- Usually bc of compensation that is on a per completion basis |
|
Leading the respondent |
Influencing the respondents answers through word voicing, voice inflection, or body language.
-Change wording |
|
Unintentional fieldwork (interviewer) error |
Interviewer commits an error while believing they are performing correctly.
- Misunderstanding, fatigue, characteristics (accent, sex, demeanor, voice, gender, lack of experience) |
|
Interviewer misunderstanding |
Interviewer believes they know how to administer a survey but does it incorrectly.
( Hard to understand instructions, gap between questionnaire designers and interviewers) |
|
Fatigue related mistakes |
Interviewer becomes tired. |
|
Intentional respondent error (2 types) |
Respondents willfully misrepresent themselves in surveys.
-Falsehoods and nonresponse |
|
Falsehoods |
Respondents fail to tell the truth in surveys.
(Could be mindless, sensitive topic, or want to deceive) |
|
Nonresponse |
Failure of prospective respondent to take part in the survey, premature termination of the interview, or refusal to answer specific questions.
(most common- distaste for surveys, busy schedules, and privacy issues)
-B2B most challenging |
|
Unintentional respondent error |
Respondent gives response that is not valid but they believe it is the truth. |
|
5 instances of unintentional respondent error |
1. misunderstanding 2. guessing 3. attention loss 4. distraction 5. fatigue |
|
Respondent misunderstanding |
Respondent gives answer without comprehending the question of instructions. |
|
Guessing |
Respondent gives answer in which they are uncertain of it's accuracy. |
|
Attention loss |
Interest in the survey wanes. |
|
Distractions |
Interruptions while taking the survey. |
|
Respondent fatigue |
Respondent becomes tired of participating in a survey. " No opinion" |
|
Supervison |
Administrators oversee the work of field data collection workers.
( may be known or disguised) |
|
Validation |
Verifies that the interviewer did the work. -aimed at falsification/cheating problem.
(Contact people to be sure they were interviewed and re administer questions) |
|
4 ways to minimize the unintentional fieldworker errors |
1. Selection and training 2. Orientation 3. Role playing 4. Methods to reduce fatigue |
|
Orientation sessions |
Meetings in which the supervisor introduces the survey and questionnaire administration requirements to fieldworkers. |
|
Role- playing sessions |
Dry runs or dress rehearsals of the questionnaire with someone plating the respondents role. |
|
5 ways to reduce intentional respondent error |
1. Anonymity 2. Confidentiality 3. Incentives 4. Validation checks 5. 3rd person techniques |
|
Anonymity |
Respondent is assured their name will not be associated with their answers. |
|
Confidentiality |
Respondent assured that their answers will remain private. |
|
Incentives |
Cash, gifts, or something of value promised to respondents in return for their participation.
(guilt trip) |
|
Validation checks |
Info provided by a respondent is confirmed during the interview. |
|
Third- person technique |
Instead of directly quizzing the respondent, the question is asked in terms of a 3rd person similar to the respondent.
(reduce falsehoods and non response) |
|
3 ways to control unintentional |
1. Questionnaire instructions and examples 2. Reversals of scale endpoints 3. Use of prompters |
|
Questionnaire instructions and examples |
Used to avoid respondent confusion, make clear. |
|
Reversals of scale endpoints |
Don't put all the negative aspects on one side and positive on the other, mix it up on a few items. - respond to individual questions. Negatively word a few questions. |
|
Prompters |
Statements that encourage people to stay on track. " We are almost done", " That was the most difficult part."
(% completed scale online) |
|
What is the marketing research industries biggest problem? |
Nonresponse |
|
Refusal |
Decline to take part in the survey.
(overcome by incentives, sponsored by edu, egoistic) |
|
Break-off |
Respondent reaches a certain point and then decides not to answer any more questions. |
|
Item ommision |
Percentage of the sample where people did not answer a specific question. " I don't know" option
(sensitive questions) |
|
Completed interview |
All the primary questions have been answered.
(secondary questions are demographic questions) - must be defined
|
|
CASRO response rate |
# of completed interviews divided by # of eligible units in sample. |
|
Dataset |
Matrix arrangement of numbers in rows and columns similar to Excel or a spreadsheet.
( row- answers, columns- questions) |
|
Data encoding |
Identification code values that are associated with the possible responses for each question on the questionnaire. |
|
Data book |
Identifies the questions,variable name/label, and codes associated.
(blank for a missing item)
|
|
Incomplete response |
Break-off where the respondent stops answering in the middle of a questionnaire. |
|
Yea and nay- saying |
Yea-saying is responding yes or strongly agree to all questions. Nay-saying is responding negative to all questions.
|
|
Middle- of- the- road pattern |
No opinion responses or "3". |