| In order to consider your surveying efforts | | | | then asked to rate their level of agreement with |
| successful, the information you collect must be | | | | a scale from 1 to 5. 1 might represent "Strongly |
| reliable and relevant to your objective. Too often, | | | | agree" while 5 represents "Strongly disagree." |
| the data that is collected from a survey is tainted. | | | | Multiple Choice |
| It may be inaccurate, vague, or lack significant | | | | Multiple choice questions are well-suited for |
| insight into the thoughts and perceptions of your | | | | uncovering the strongest reason for a given |
| population. More often than not, you can control | | | | feeling. Typically, a number of choices are made |
| these issues. The manner in which you create | | | | available from which your participant chooses. |
| your survey questions and how you combine a | | | | Even if more than one answer is accurate, the |
| variety of different types can help you elicit the | | | | respondent is forced to choose the strongest |
| information you need to take action. | | | | answer. |
| Today, I'm going to explain the fundamentals of a | | | | For example, he or she may be asked, "Why do |
| good survey question. I'll describe the variations | | | | you enjoy your job?" In most cases, there will |
| you can use, the proper way to construct them, | | | | likely be several possible reasons (for example, |
| and some factors to consider to ensure your | | | | high salary, small commute, intellectual challenge, |
| results are useful. | | | | etc.). By forcing that person to select only one |
| Closed-Ended Versus Open-Ended | | | | answer, you can determine the most important |
| Almost all questions that you can design for your | | | | reason. |
| surveys will either be closed-ended or open-ended. | | | | Priority |
| Let me take a moment to define the two types. | | | | Queries which ask respondents to prioritize the |
| Closed-ended queries are those that only let a | | | | factors that are important to them can be |
| person to choose from a given set of responses. | | | | especially useful in identifying behaviors, marketing |
| One of those responses might be "Other," but | | | | opportunities, and other valuable data. A number |
| even then, the answer is limited. By contrast, | | | | of choices (usually, 5) are presented and the |
| open-ended queries allow respondents to | | | | participant is asked to rate them from 1 to 5. |
| elaborate. | | | | For example, they might be asked, "Rate the |
| One of the advantages of a closed-ended | | | | following items in order of importance with |
| question is that the responses are easy to | | | | regards to how you would choose a health club |
| categorize. They're in a standard format, making | | | | membership." The items might include monthly |
| them especially well-suited for statistical analysis. | | | | fees, flexibility of terms, and availability of the |
| The drawback is that this type of query can be | | | | trainers. This type of survey question will help you |
| deceptively complicated to create. Open-ended | | | | identify areas that are most important to your |
| questions can often reveal insightful information. | | | | respondents in relation to other areas. |
| Their inherent disadvantage is that they're | | | | Critical Factors In Query Design |
| unsuitable for statistical analysis. In fact, | | | | The keys to obtaining useful data from your |
| interpreting the responses can often be | | | | surveys is to focus upon one element, try to |
| challenging. | | | | accommodate all potential responses, and |
| Scale Queries | | | | encourage your respondents to answer truthfully. |
| Scale queries are most appropriate when you | | | | Today, we've focused most of our attention on |
| want to determine your survey population's | | | | closed-ended queries. They're a bit more difficult |
| feelings toward a specific topic. Normally, | | | | to design because they must adhere to a specific |
| respondents are asked to choose a rating | | | | structure. Open-ended queries are far easier to |
| between 1 and 5 that most closely represents | | | | create because they allow broader answers. |
| their personal viewpoint. For example, the question | | | | Ideally, your surveys should blend a mix of each |
| - or statement - might be "A college education is | | | | type that I've profiled above. |
| critical for future success." Your respondent is | | | | |