Basic Consideration
You have to make sure that your questionnaire questions really reflect the original research question.
It is also helpful, before you start formulating the questions, to study the characteristics of your target population. This would allow you to use a language that they can understand.
Another very important aspect of Survey Research is the different types of data that the questions generate. This is extremely important when we decide the statistical procedure to analyze the data.
The technical term to refer to the different types of data is “levels of measurement”.
This is NOT a course on statistics. Still, it is necessary to understand the different levels of measurement in order to choose the correct type of analysis for our data.
The four level of measurement are:
- nominal,
- ordinal,
- interval,
- and ratio
Nominal data do not have any numeric value. When you ask in an interview about the gender of the respondent, the fact that the identify them as male or female (or any other option you add to your questionnaire), does not give them any numeric value. Female is not higher, bigger, or larger than male. Even when we attribute numeric values to nominal variables, this value is completely arbitrary. In the coding process, we can assign 1 to male and 2 to female, or the other way around. That will not change anything in our outcomes.
Ordinal data can be ordered on the basis of rank. In the case, the distance between the different values has no meaning, or it is unknown. You answer to ordinal question, for instance, in customer satisfaction surveys. When, after having bought any product or services thay ask you to rate the experience as
1 (extremely unsatisfactory), 2, (unsatisfactory), 3 (neutral), 4 (Satisfactory), 5 (extremely satisfactory)
In the case of the interval data, the distance between the values does have a meaning. This is the scale used, for instance, to measure temperature. The difference between 30 and 32 degrees is the same as the difference between 40 and 42 degrees. Mathematical computations are limited with interval data. Only the mean is used in this measurement level.
In the ratio measurement level, the difference between the values also has a meaning. In addition to that, ratio scales have an absolute “0” value, allow all mathematical computations and can be converted into other units. When you are measuring the number of clients of a company, you are using ratio scales. Time or age can also be used as ratio scales.
We will not elaborate on this subject. Still, if you are interested in exploring it, you can go this this site.
Types of Questions
The “Dichotomous” Question
Allows only two possible options (yes/no)
| Have you ever smoked marihuana? |
|---|
| Yes |
| No |
True/false version:
Sometimes they offer a third option to avoid taking position on an issue.
| Blond women are dumb. |
|---|
| True |
| False |
| Impossible to say |
Multiple Choice
In the classic multiple choice question, you give the interviewee different options to choose his/her answer.
| How did you hear about CCSU? |
|---|
| Television |
| Radio |
| Newspapers ads |
| Word-of-mouth |
| Internet |
| Other. Please, specify |
Categorical Question
Modality of multiple choice in which the respondent must select one category to identify him/herself.
| What is your gender? |
|---|
| Male |
| Female |
| Other |
| What is your current marital status? | is | your | current | marital | status? |
|---|---|---|---|---|---|
| Single | |||||
| Married | |||||
| Divorced | |||||
| Separated | |||||
| Widowed |
Make sure that the different categories are mutually exclusive
| Were are you living? |
|---|
| House |
| Apartment |
| Suburbs |
You can live, at the same time, in a house and the suburbs.
Ordinal Question
Ordinal questions allow you to establish the priorities and preferences of you target population.
| Please rank the importance of the following qualities in a college professor. (Please fill in your rank order in the spaces provided using the numbers 1 through 5) |
|---|
| A professor that is a fair grader |
| A professor who is physically attractive |
| A professor who is intellectually stimulating |
| A professor who is a disciplinarian |
| A professor who is a strong motivator |
| How would you rank following brands according to their reliability? (Please fill in your rank order in the spaces provided using the numbers 1 through 10) |
|---|
| Ford |
| Toyota |
| Honda |
| Buick |
| Cadillac |
| Jeep |
| Volvo |
| Chrysler |
| Saab |
| BMW |
Numerical Question
The answer must be, in this case, a number.
| What is your yearly income? |
|---|
| Less than $25,000 |
| $25,000 to $50,000 |
| $50,000 to $75,000 |
| $75,000 to $100,000 |
| $100,000 or more |
| What is your current age? |
|---|
| Less than 18 |
| 18 to 29 |
| 30 to 39 |
| 40 to 49 |
| 50 or older |
Scales
Scales allow us to explore the attitudes and feelings of our target population.
Rating Scale
Likert Scale
Semantic Differential Scale
Stapel Scale
Rating Scale
Respondent can rate the quality of something – anything.
| Which of the following categories best describes you experience in the course Public Opinion? | ||||
|---|---|---|---|---|
| Very unpleasant | Somewhat unpleasant | Neither pleasant nor unpleasant | somewhat pleasant | Very pleasant |
Likert Scale
The most popular scale to measure attitudes.
| I think the instructor of this course is fantastic! | ||||
|---|---|---|---|---|
| Strongly disagree | Disagree | Neutral | Agree | Strongly agree |
Semantic Differential Scale
The Semantic Differential scale allows the respondents to describe their feelings or attitudes toward any given topic based on a 10 point scale that has bi-polar adjectives at each end.
How would you describe you experience in this course?
Very bad
| Bad | Good | ||||||||
| 10 | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | 1 |
How would you rate the food at CCSU’s canteen?
| deplorable | Excellent | ||||||||
| 10 | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | 1 |
How do you feel about learning foreign languages in higher education?
| Unimportant | Important | ||||||||
| 10 | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | 1 |
How would you rate the price of the iPhone?
| Cheap | Expensive | ||||||||
| 10 | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | 1 |
How do you feel about your recent internship experience?
| Useless | Useful | ||||||||
| 10 | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | 1 |
Stapel Scale
Unipolar Scale with no zero value. It indicates how well an adjective describes something (an issue, a person, a product, …)
Do you think that the word “modern” describes the program of the communication department?
- (-5) Poorly describe
- (-4)
- (-3)
- (-2)
- (-1)
- Modern
- (+1)
- (+2)
- (+3)
- (+4)
- (+5) Describes very well
Matrix
The matrix allows you to compare different items based on one specific criteria.
How hot are, in your opinion, following professors at the Communication Department?
| cold | cool | warm | hot | burning | |
| Prof. del Ama | |||||
| Prof. Mendez | |||||
| Prof. White | |||||
| Prof. Tyson | |||||
| Prof. Ju |
Open Ended Questions
Open ended questions give you the opportunity to go into depth in some aspects of any given issue. They are time consuming because it is a tiresome process to identify logical categories.
Tell me about you relationship with the instructor.
What do you think about the democratic candidate for the next elections in Connecticut?
How do you feel about the proposed health reform?
What kind of challenges did you face in this course?
Demographic Questions
They provide you with necessary information to analyze the data.
age,
gender,
income,
race,
ethnicity,
marital status,
educational level,
geographic place of residence,
number of children (siblings), …
I have tried to reduce this survey on questions to the most common types, the ones you may use in your projects.
I use this platform to direct you to other Web-sites where you can keep exploring the subject.
In this this link, you can find a more detailed review of survey questions.