Class-6 (UNK)

Class – 6

Editing & Coding

Open ended questions on a questionnaire (Quantitative) should be bare minimum as it is difficult to code and analyze it as against a closed ended question.

Editing and Coding (Part of Data Processing)

Question - What will a researcher say in case he / she has a presentation with client tomorrow and the data which has just arrived in evening a day before is not having answers to certain questions in a questionnaire?

Answer – Sir, the data just arrived yesterday, we are looking into it and examining it, but we have found few errors / glitches in the data. While most of the presentation is fine, we just need to correct / confirm a few things and hence, need a day or 2 extra just to confirm it. So, I request 2 options – first option is we go ahead with the presentation and come back and present this part at another date or second option is we defer the entire presentation by 2 days. What do you want me to do?

It is better to get this sorted out upfront rather than going into the presentation and embarrassing the client over there. Integrity and transparency in such scenarios is appreciated by clients.

About editing - The agency should have in-house editing department.

Why do we do editing?

· Consistency – Answers are consistent with the question asked and its option given (Respondent must select an option given in the questionnaire for a closed ended question).

· Completeness – No question should be left unanswered

· Questions answered out of order

Don’t Know –

· Legitimate Don’t Know – The person has not previously experienced / has no knowledge of the subject being discussed, hence a legitimate don’t know. Example – How is RM2 class. Answer is Legitimate Don’t know as the class has not yet started.

· Reluctant Don’t Know (Negative perception, like always choosing center point on an odd point likert scale)– Someone just gave an exam and is asked about how was the exam just after he / she comes out of the examination hall and he / she replies – I do not know, is an example of reluctant Don’t know. Another example could be professor of RM asking – How is RM1 class? Answer is – Reluctant Don’t know as the student does not want to give a negative opinion.

· Confused Don’t Know – Example – How is RM class? Respondent is confused as he does not know which RM is being talked about.

Legitimate Don’t know is removed from the sample, whereas Reluctant Don’t know and confused Don’t know is presented to client because they require different kinds of actions.

In case someone has a contract with a company of not to speak out on certain topics and because of the same he / she says don’t know, then that don’t know is neither of the above.

Every question must have a numerical code because it is going to be used for coding purposed (using software like SPSS).

Coding of Closed Ended Questions

About Awareness Question coding sheet preparation – (1^st mentioned, Other mentioned, Aided) –

One column to be left for “1^st-mentioned” in coding excel (as only 1 answer could be the “1^st-mentioned”). (N-1) columns for “other-mentioned” - it is the maximum number of columns for “other-mentioned”. N stands for number of brands being present in awareness question. (N-1) for “aided” column also (maximum columns) in case “1^st-mentioned” is present and nothing is present in “other-mentioned”. (Assumption – There is no “Others (specify)” option along with various brands).

“Others (specify)” option in “1^st-mentioned” column can’t be more than 5% of total sample. Else the researcher is not having good knowledge / has not done proper piloting before preparing the questionnaire.

In case nothing is mentioned in the 1^st mentioned, the sample selected is wrong (Hence, nothing is present in other-mentioned and everything gets recorded in aided which will be equal to “N”).

Brand Name	First Mentioned	Other Mentioned	Aided
Galaxy	1	1	1
Galaxy Jewels	2	2	2
Cadburys	3	3	3
Kit Kat	4	4	4
Hersheys	5	5	5
Lint	6	6	6
Mars	7	7	7
Tiffany	8	8	8
Others (Specify)	9	9	9

Coding of open ended question

Question - Why do you eat chocolate?

· Respondent 1 – Like, tastes good

· Respondent 2 – Like, taste, pleasure, joy, happiness

· Respondent 3 - Lousy, Feel good factor

· Respondent 4 – Taste good, Desert (after dinner)

· Respondent 5 – As Snack

Respondent 1 response – like & taste good is called VERBATUM(responses recorded as it is verbally spoken by respondent)

Coding starts –

· Like means taste good and sweet is a taste, hence, like, sweet and taste are Code 1 (4 respondents out of 5)

· Pleasure / joy / happiness / Feel Good / mood up-lifter – Code 2 (2 respondents out of 5)

· As a Desert – Code 3 (1 respondent out of 5)

· As a snack – Code 4 (1 respondent out of 5)

(1 respondent out of 5) is called TABULATIONS

Like means taste good, interpretation and clubbing into similar characteristic features is called CODE FRAMES (Grouping similar features)

Conclusion –

Majority of respondents eat a chocolate because of its taste. 40% respondent eats chocolate because of the pleasure / joy they get by eating it (Mood up-lifter).

Coding can only start (i.e. preparation of tabulations and code frames) after 70% sample size is achieved. Hence, coding of open ended question is cumbersome / complex / subjective as every questionnaire is read and then categorized (grouping of different options can be questioned by client) and tabulated, unlike a closed ended question (where option is selected/ clearly). Hence, open ended question is not encouraged by the researcher.

Data Analysis –

3 broad categories of data-analysis –

· Uni-variate (Analysis of one variable)

· Bi-variate (Analysis of 2 variables)

· Multi-variate (Analysis of more than 2 variables)

Advantage of DU over Etisalat – Du charges by Pulse rate and not by the minute rate. Network coverage and clarity coverage is better for Etisalat as compared to Du. Similarly, Window is more compatible but Mac is a superior product. Hence, in creative fields, Mac is used as against Windows. Hence, in advertising field, it is all Mac.

How long the questionnaire should be

It depends on the following 2 assumptions –

1. It is a face to face interview where the interviewer is having the questionnaire and he/she is asking the respondent and filling the responses by himself / her-self based on responses received.

2. The length of the survey is 20 minutes (It is done in order to ensure that maximum interview time is not exceeding 30 minutes in case respondent takes more time to answer every question. Research shows that fatigue to the respondent happens if the interview is more than 40 minutes and they might become jittery)

Time spent to conduct an interview (quantitative – including both close ended and open ended)

· Face to face interview – 25 to 30 minutes

· Telephonic interview – 15 to 20 minutes

Respondent is not paid in any of the above cases (quantitative), client is only paying the researcher. But in case of focus group discussion (qualitative), the respondent gets paid / gets gift. Also, they get paid for mystery shopping.

………………………………………………………………………………………………………………………………………………………………….

Types of Questions asked during Market Research

Closed ended questions

Name	Description	Example
Dichotomous	A question with 2 possible answers	In arranging this trip, did you personally phone Air India (Yes-1 / No-2)
Multiple choice	A question with 3 or more answers	With whom are you travelling tonight (No one-1 / Spouse-2 / Children-3 / Spouse and children-4 / Friends & relatives-5 / Business tour group-6)
Likert Scale	A statement with which one shows amount of agreement or disagreement	Small airlines generally give better services than large ones (Strongly disagree-1 / Disagree-2 / Agree – 3 / Strongly Agree – 4)
Semantic differential scale	A scale connecting 2 bipolar words. The respondent selects the point that represents his / her opinion	Air India is Large……………………….Small Experienced……………………..Inexperienced Modern……………………….….Old-fashioned
Rating scale	A scale that rates some attributes from “poor” to “excellent”	Food services of Air India is (Excellent – 1 / Very Good – 2 / Good – 3 / Fair – 4 / Poor – 5)
Importance scale	A€ scale that rates the importance of some attributes	Airline food service to me is (Extremely important – 1/ Very important -2 / Somewhat important-3 / Not very important – 4/ Not at all important – 5)
Intention to Buy scale	A scale that describes respondent’s intention to buy	If an in-flight telephone were available on a long flight, I would (Definitely buy-1 / Probably buy-2 / Probably not buy-3 / Definitely not buy-4)

Open Ended Questions

Name	Description	Example
Completely unstructured	A question that respondents can answer in an almost unlimited number of ways	What is your opinion about Air India
Sentence completion	Words are presented, one at a time, and respondents mention the 1st word that comes to their mind	What is the first word that comes to your mind when you hear: Sweet (Chocolate, sugar, sweet, colas)
Story completion	An incomplete story is presented and respondents are asked to complete it	When I choose an airline, the most important consideration in my decision is _______________
Picture	A picture of 2 characters is presented, with one making a statement. Respondents are asked to identify with the other and fill in the empty position
Thematic Apperception Test (TAT)	A picture is presented and respondents are asked to make a story about what they think is happening or may happen in the picture

Questionnaire DOs and DON’Ts

1. Ensure that questions are asked without bias. Don’t lead the respondent into an answer.

2. Make the questions as simple as possible. Questions that include multiple ideas or two questions in one will confuse respondents.

3. Make the questions specific. Sometimes it’s advisable to add memory cues. For example, be specific with time periods.

4. Avoid jargon or shortlisted. Avoid trade jargons, acronyms, and initials not in everyday use.

5. Steer clear of sophisticated or uncommon words. Only use of words in common speech.

6. Avoid ambiguous words. Words such as “usually” or “frequently” have no specific meaning.

7. Avoid questions with a negative in them. It is better to say, ”Do you ever……..?” than “Do you never…..?”

8. Avoid hypothetical questions. It’s difficult to answer questions about imaginary situations. Answers aren’t necessarily reliable.

9. Do not use words that could be misheard. This is especially important when administering the interview over the telephone. “What is your opinion of sects?” could yield interesting but not necessarily relevant answers.

10. Desensitize questions by using response bands. To ask people their age or ask companies about employee turnover rates, offer a range of response bands instead of precise numbers.

11. Ensure that fixed responses do not overlap. Categories used in fixed-response questions should be distinct and not overlap.

12. Allow for “other” in fixed-response questions. Prescribed answers should always allow for a response other than those listed.

13. Surveys

Characteristic	Telephone Surveys	Door-to-door Personal Interview	Mall intercept Personal Interview	Mail Surveys
Speed of data collection	Very fast	Moderate to fast	Fast	Slow (Researcher has no control over return of questionnaire)
Geographical flexibility	High	Limited to moderate	Confined, urban bias	High
Respondent Cooperation	Good	Excellent	Moderate to low	Moderate (Poorly design questionnaire will have low response rate)
Versatility of questioning	Moderate	Quite versatile	Extremely versatile	High standardized format
Questionnaire length	Moderate	Long	Moderate to long	Varies depending on incentive
Item non-response		Low	Medium	High
Possibility of respondent misunderstanding	Average	Lowest	Lowest	Highest – no interviewer present for clarification
Degree of interviewer influence answer	Moderate	High	Highest	None – interviewer absent
Supervision of Interviews	High, especially with central locations	Moderate	Moderate to high	Not applicable
Anonymity of the respondent	Moderate	Low	Low	High
Ease of call back or follow-up	Easy	Difficult	Difficult	Easy, but takes time
Cost	Low to moderate	Highest	Moderate to High	Lowest
Special features	Fieldwork and supervision of data collection are simplified; quite adaptable to computer technology	Visual materials may be shown or demonstrated; extended probing possible	Taste test, viewing of TV commercials possible	NA

Characteristic	E-mail questionnaire	Internet Surveys
Speed of data collection	Instantaneous	Instantaneous
Geographical flexibility	Worldwide	Worldwide
Respondent Cooperation	Varies depending if e-mail is seen as “spam”	Depends on speed of internet, respondents computer skills
Possibility of respondent misunderstanding		High
Degree of interviewer influence answer		None
Supervision of Interviews		Not required
Anonymity of the respondent		Respondent can be anonymous or unknown
Ease of call back or follow-up		Difficult unless e-mail address is known
Cost	Cheap distribution and processing cost	Cost effective
Special features	Flexible, extensive differences in capabilities of respondents computers and e-mail software limit the types of questions and the layout	Allow graphics and streamlining media

Characteristics of a Focus Group discussion

Ø Group consists of 8-10 people for the interview

Ø The Interview starts with a broad topic & focus in on specific issues

Ø The interview is unstructured and free flowing (no structured questionnaire)

Ø Moderator (interviewer in a focus group) develops rapport, stimulates spontaneous response, interacts, listens, controls and gives a chance to everyone to express their opinion

The focus group members have same demographics, lifestyle and attitudes