| Argument open-ended question |
An open-ended question that asks the participant about both sides of an issue. |
| Closed-ended question |
A question that offers a list of response options. |
| Contributor |
Anyone who significantly contributed to the purpose, method, or analysis of a survey and could speak about the survey if asked. |
| Convenience sample |
A sample selected based on the convenience of contacting the participants. |
| Cover letter |
A letter included in every survey that answers the following questions:
 Who is conducting the survey?
Why is the survey being done?
What information is being sought?
How have the participants been selected?
What will be done with the results?
Are the answers confidential?
By when must the survey be completed?
Will participants get information on the survey results? |
| Dichotomous item |
A form of nominal closed-ended question. Dichotomous items allow the participant to select one of two possible responses. |
| Frequency distribution |
A statistic that identifies the number of times each response appears in the overall data from a question. |
| Focus group |
Administration of a survey in a group setting. Participants can share and discuss their opinions. A trained facilitator conducts the session. |
| Funneling technique |
Beginning a section of a survey with an open-ended question, then moving into closed-ended questions about very specific topics. |
| Group |
Also referred to as a population. |
| Group-administered survey |
Administration of a survey to a group of participants, each of whom completes his or her own survey at the same time. The business analyst supervises and answers questions for clarification. |
| Incorrect data |
Data that do not make sense based on the question asked. |
| Interval closed-ended question |
A closed-ended question with response options that are presented in a logical order. There are equal differences between each adjoining response category for interval closed-ended questions. |
| Interviewee |
A participant in a one-on-one structured interview setting. |
| Loaded word |
A word that can cause a participant to have immediate positive or negative feelings. |
| Mail-out survey |
A survey that is sent to participants. There is no contact between participants and the business analyst. Forms of mail-out surveys include paper mail, email, and online. |
| Mean |
The average of a set of scores. |
| Meaningless responses |
The results from a question that can be answered without any knowledge of the topic. |
| Median |
The point that cuts the distribution of responses in half. The median is sometimes considered the “typical” response. |
| Missing data |
The absence of data if not all participants answer a particular question. |
| Mode |
The most frequently occurring score in a distribution. |
| Multiple-response item |
A form of nominal closed-ended question that allows the participant to select one or more options from a list. Multiple-response items may also be called multiple-choice questions, categorical response items, or checklists. |
| Nominal closed-ended question |
A type of closed-ended question that is generally used to gather factual information. Nominal closed-ended questions are not used to measure amounts. |
| One-on-one interview |
An in-person interview or face-to-face interview. |
| One-on-one survey |
A survey where there is a well-defined list of questions to be asked in each one-on-one survey and a specific order. Questions are not dropped or added from one survey to the next. |
| Open-ended question |
A question that allows participants to answer in their own words. A wide range of responses are possible, and answers are usually longer than one word. |
| Ordinal closed-ended question |
A question that allows the participant to designate the place of an option in an ordered sequence. |
| Outlying data |
Responses that are not consistent with the rest of the data. Also called outliers. |
| Population |
A group of people who have common qualities or characteristics. |
| Power User |
A system user who has the ability to use advanced features of the program that are beyond the abilities of a normal user. A power user is not, however, a programmer and may or may not be a system administrator. |
| Random sampling |
Selecting every nth name from a list of potential participants. Also known as systematic sampling. |
| Range |
The distance between the highest and lowest scores in a distribution. |
| Ranking-order scale question |
A type of ordinal closed-ended question used to gather facts, opinions, attitudes, or judgments. There are equal differences between each adjacent response category for ranking-order scale questions. |
| Rating-scale question |
A question that requires the participant to assign a value to something, but unlike ranking-order scales, the participant is not required to compare one option with other options in the same question. |
| Recruited sample |
A sample composed of participants recruited by phone, email, or in person. |
| Response rate |
The percentage of participants invited to participate in a survey that actually completes the survey. |
| Sample |
The portion of the population included in a survey. |
| Self-administered survey |
Surveying method in which a business analyst meets with participants, but participants fill out the survey by themselves, with or without the business analyst present. |
| Single-response item |
An open-ended question with a blank for the response. |
| Snowball sample |
Sampling technique in which a few participants are contacted and asked to provide referrals to other potential participants. |
| Skip pattern question |
A question that guides the participant to do one of two things based on his or her response. |
| Stakeholder Stakeholder analysis |
A person or group with an interest in the success of a survey. A technique used to ensure that all interested parties are identified and their needs, wants, and expectations are understood. |
| Survey |
“A means of eliciting information from many people, anonymously, in a relatively short time” (IIBA® 2006, 177). |
| Systematic sampling |
Selecting every nth name from a list of potential participants. Also called random sampling. |
| Telephone survey |
A survey administered by phone; also known as a telephone poll. A telephone survey should be no more than 15 minutes in length. |
| Transition |
Language used in a survey to show participants that the survey is changing direction. |
| Unrestricted sampling |
Allowing anyone who finds the survey—for example, on the Internet—to participate. |
