How to Write Institution Specific Questions (ISQs)

This article details how to write useful Institution Specific Questions (ISQs).

What information do you need to make a decision? Clearly define the information you need to collect, make it ‘need to know’ and not just ‘want to know’.

Consider if the respondents will willingly provide the information and if they have the information to answer the questions.

Don’t try to cover too much in one question: Questions with compound clauses may cause confusion and irritation giving the respondent a reason to skip the question or stop taking the survey.
- Example: “How satisfied are you with the temperature and flavor of your coffee?”
- Why this is wrong: A respondent can like the temperature but not the flavor so they have no way of answering this question as it’s written.
- How to fix: Must use two questions “How satisfied are you with the temperature of your coffee?” and “How satisfied are you with the flavor of your coffee?”

Don’t use words that could have more than one meaning: If it is possible for respondents to interpret questions in dissimilar terms, they will.
- Example: “How compatible are you and your roommate?”
- Why this is wrong: One respondent may interpret “compatible” to mean there are no huge arguments or problems while another respondent may interpret “compatible” to mean a level of respect and admiration.
- How to fix: Use a couple of questions like “To what degree do you and roommate easily resolve conflicts?” and “To what degree do you and roommate enjoy spending time with each other?”

Don’t use multiple negatives: It introduces undue complexity and causes stress to the respondent.
- Example: “To what degree do you not like to attend class?”
- Why this is wrong: Respondents have to try and figure out how to answer this question with the response key.
- How to fix: “To what degree do you enjoy attending class?”

Don’t ask a question that leads the respondent to give a particular response: If a negative response to a question appears to be mean-spirited, it may bias the survey toward too many affirmative responses.
- Example: “To what degree do you feel you’re a racist?”
- Why this is wrong: The question is asked with a hostile tone and the respondent will automatically answer “not at all” instead of considering the question.
- How to fix: Consider a series of questions like “To what degree do you enjoy spending time with people who are of a different race than you?”

Don’t antagonize the respondent by using hostile language: Don’t use possible sexist or racist wordings, “loaded” terms and avoid the use of negative, inflammatory words like allege, blame, claim, demand, fault, failure, forbid, ill-advised, ineptness, misinformed, must, neglected, questionable, rigid, unreasonable

Don’t make your questions general – be specific: Ask about specifics (performance, availability, ease of use, etc.)
- Example: “How satisfied are you with the dining facility?”
- Why this is wrong: Satisfaction with a dining facility can be impacted by a lot of variables like type of food, quality of food, food preparation, dining environment, staff, etc. You need to be specific regarding what area of satisfaction you’re measuring.
- How to fix: Consider a series of questions like “How satisfied are you with the type of menu items offered?” and “How satisfied are you with the preparation of food?” etc.

Don’t use time frames that are too long: People’s ability to recall the past is limited, the more current and specific the question, the better.
- Example: “Have you attended a concert on campus?”
- Why this is wrong: Some respondents may have attended a concert years ago but it’s their current behavior that should be under consideration.
- How to fix: “Have you attended a concert on campus during the last school term?”

Do keep things simple, concise and clear: Avoid generalities such as, “How popular is AI at this time?” How popular with whom and compared to whom?

Do start with a question that is comfortable for the respondent to answer: It will help the respondent establish focus.
- Example: “How satisfied are you with the choice of student activities offered?”

Do make your questions understandable to your targeted population: Consider whether a formal approach or a more personalized style will produce better results, but avoid abbreviations, technical terms and jargon.

Do make your questions relevant, and consistent with the survey goals: People’s ability to recall the past is limited, the more current and specific the question, the better.

Utilizing a person who is trained in survey development to review your questions after you’ve written them. Chances are, there are people on your campus that can help. Contact your institutional research office or assessment office.