Chapter 10

Surveys

In This Chapter:

A survey is a process for quickly gathering information without any type of verification. The term survey can also be used to describe an activity such as a review of customer support problems or product failure data. This information is already available to the organization and simply needs to be analyzed. Surveys that gather information without verification are used to:

Understand the process undergoing change.

Identify significant areas warranting future study or analysis.

Obtain information to help determine the appropriate requirements elicitation and analysis activities.

Types of Requirements Elicitation Surveys

A well-crafted survey asks questions in a variety of ways. As with interview questions, there are two basic types of survey questions from which to choose, open-ended and closed-ended.

Open-ended questions do not have one right answer, but rather give respondents an opportunity to answer in their own words. While the responses to open-ended questions can be quite useful, they are more difficult to interpret and catalogue.

Closed-ended questions have a finite set of answers for the respondent to select. Closed-ended questions are easy to standardize and the data lends itself to statistical analysis. Closed-ended questions are difficult to formulate so that all possible answers are provided. To get around this, there is often a choice labeled “Other.” If a respondent chooses this option, it is helpful to provide space for an explanation. There are several types of closed-ended questions:

Likert scale allows for a ranking from “not very important” to “extremely important” or from “strongly disagree” to “strongly agree.”

Multiple choice allows the respondents to select the best answer from among possible options.

Ordinal asks the respondent to rank order a list of items.

Categorical allows the respondent to select a category.

Numerical asks for a real number for the answer.

Benefits of Requirements Elicitation Surveys

Surveys are administered to groups of stakeholders to determine information about customers, work practices, and attitudes. Responses are analyzed for functional and supplemental requirements and stakeholder interests and positions. Surveys are particularly valuable when customers are dispersed or employees are in multiple locations. Table 10-1 shows the benefits versus the disadvantages of surveys.

Table 10-1—Benefits versus Disadvantages of Surveys

Rules for Effective Requirements Elicitation Surveys

The business analyst should keep several rules in mind when conducting surveys:

Understand what information is needed from the survey respondents. For example:

Focus on the high-priority risks that have been identified in early project meetings or interviews, and write questions to confirm the probability or impact of the risk event or issue.

Identify user satisfaction levels with the existing systems to create a baseline for improvements.

Create clear and concise questions. Constructing a valid survey instrument is challenging. Strive to:

Be direct. “How many hours do you use the system” versus “How often do you use the system?”

Be unambiguous. “Rate yourself as being a novice, an average user, or an expert user of the system” versus “Do you consider yourself to be an expert?”

Save complex questions for later in the survey. “In the space below, provide additional information about how the system could improve the processing of exempt payroll.”

Save demographic information—age group, role, division, gender—for last to avoid feelings that this is a personal reflection.

Create rewards for participating. For example:

Provide visibility for the division or department with the highest percentage of responses.

Consider a monetary incentive or reward a random participant with a voucher for a night on the town.

Create the survey using the inexpensive online tools that are readily available. It is increasingly easy and cost-effective to develop surveys online and to create an anonymous portal that calculates results.

Notify the participants when the survey is available and continue to remind them to participate.

Analyze the survey results. This is the hardest part. There may be some poorly performing questions due to ambiguity or confusion or even errors in the survey instrument itself.

Tailoring Requirements Elicitation Surveys

Surveys can be valuable for most projects. The business analyst scales survey activities to the project environment. Table 10-2 shows interview variances scaled to the project profile.

Table 10-2—Surveys Scaled to Project Size, Risk, Complexity