MDM_collection_questions

__Types of Survey Questions__  There are 2 categories into which all survey questions fall: **open-ended** and **closed-ended** questions. __Open-Ended Questions__
 * Ask for descriptions or more detailed responses than closed-ended questions
 * i.e. "Where is your favourite place to vacation, and why?"
 * Allow the respondent more freedom to express opinions and feelings about a question
 * Much more difficult to analyze and input as data then closed-ended questions

__Closed-Ended Questions__
 * Much more common than open-ended questions
 * Finite set of responses the respondent must choose from
 * Higher risk of non-response bias if the options do not include every possible situation
 * Closed-ended questions are harder to write than open-ended questions because of the need to foresee every response
 * Provide data that is very easy to quickly analyze and interpret

__Types of Closed-Ended Questions__ There are 5 types of closed-ended questions, and each is useful for different things, listed below. When there is an "other" option in a closed-ended question, you must leave a space for the respondent to write their "other" response.

1. __Likert Scale Questions__ Likert Scale questions are used when you want to find the feelings of the respondent; the respondent must choose a number on a scale to indicate how closely their feelings match the question. Make sure to indicate what the highest and lowest numbers mean. //example: "How satisfied are you with the customer service at this restaurant? Please circle a number from 1 to 5, where 1 is not satisfied at all, and 5 is very satisfied." 1 2 3 4 5//

2. __Multiple Choice Questions__ Multiple choice questions are the easiest to analyze. In a multiple choice question, there are a finite number of answers that the respondent can choose from. The respondent chooses the best option or options, depending on the instructions. Always include instructions on how many answers the respondent may choose. //example: "What do you do in your spare time? Please circle all that apply." a. Read b. Sports c. Sleep d. Other//

3. __Ordinal Questions__ Ordinal questions ask the respondent to rank a number of options. Do not exceed 5 options, as the respondent may find it difficult to rank more than 5 options, and the data produced may be unreliable. These questions can also be confusing if they are not written properly; for example, the respondent may be confused as to whether they use each number only once, or if they can use them more than once. If each number can only be used once this will also skew the rating of the others if one of the respondents does not participate in/care about one or more of the options. //example: "Please rank the following options from 1 to 5, with 1 being the least important and 5 being the most important. Please use each number only once."// ---//Swimming// ---//Fishing ---Sports ---School --Shopping//

4. __Categorical Questions__ Categorical questions are used when the respondent must fit into one category, and no more. These are generally used to determine age, gender, etc. //example: "Please circle your sex." M F "Please circle your age group." 13 and under 14-19 20-29 30-39 40-59 60+ // 

// 5. // __Numerical Questions__ Numerical questions are used to determine real numbers, such as date of birth, health card number, money, etc. They are used as a quantitative evaluation of something. //example: "How many pets do you own?"

__How to Write Good Survey Questions__ // When writing survey questions, the top priority is to always remain objective. Listed below are some other things you should look out for:
 * Do not make questions that may be ambiguous - ensure that they are written clearly, and easy to understand
 * Do not jump from subject to subject; make questions flow
 * Do not make any assumptions, i.e. asking something like "How satisfied are you with your home insurance?" assumes the respondent owns a home and has home insurance
 * Do not use abbreviations
 * State your confidentiality policy first, so that people feel safer answering honestly, especially if the survey deals with sensitive material
 * Avoid ordinal questions, which are difficult to make and answer - data will not be nearly as accurate as with other types of questions
 * Be as objective as possible and do not make questions that lead a person to a certain response
 * When asking questions, they should not rely on responses to other questions in the survey
 * Use common words
 * Keep lists or multiple choice options short, but ensure that every response possible is included
 * If there is an "other" category, leave a space for the respondent to write their response
 * Avoid using negatives in questions, as responding "no" to these will create a double negative, and confuse the respondent
 * Make sure questions do not ask for things that need to be looked up, or things that stretch the memory
 * When using a likert scale, always make 1 the least and 5 the most of whatever you are measuring, and state this in the question
 * State clear instructions for multiple choice questions about whether one or more options may be circled
 * Before writing any questions, figure out what it is you want to find out with your survey, and write the questions intelligently, so that you do not have to repeat the survey / miss crucial information / word the questions wrong and render the data useless

BY MICHAEL AND HEATHER

 Waddington, H. (2000) Types of Survey Questions. In B. Hoffman (Ed.), //Encyclopedia of Educational Technology.// Retrieved October 27, 2009, from __http://coe.sdsu.edu/eet/Articles/surveyquest/start.htm__ **
 * Bibliography:

 How To Write A Good Survey. (1997) Retrieved October 23, 2009, from __http://www.accesscable.net/~infopoll/tips.htm__