Chapter 8: surveys and questionarres
closed
When respondents are asked a question (or given a statement) and then given a set of responses to select from, the question is ....... Questions that require a simple "yes" or "no" answer are .....questions. Likewise, questions that ask for a specific piece of information are ..... For example, "How old are you?" is also a .... question be- cause simply answering "I'm 23" answers the question.
open
When respondents use their own words to respond to a question, it is an ...... question. This question form gives you data from a respondent's point of view rather than your own . Responses to ..... questions are difficult to translate into numerical equivalents for use in statistical tests. For that reason, .....questions are used sparingly, and often at the end of the survey after the closed-ended questions.
cross-sectional
Whether a questionnaire, survey, or poll, there are four commonalities for this method of quantitative data collection. First, each is part of a fixed design; that is, the design of the questionnaire, survey, or poll does not change; it is standardized. All participants are asked the same questions and asked to respond on the same scales or categories. Second, all three forms of data col- lection help researchers collect data from a large number of people. Third, data are collected from a sample of individuals from an identified population. Finally, the data are ....... That is, data are treated as if all data were collected at the same time, even if takes a few days or weeks to collect the data.
social desirability
the presence of the interviewer may create a ..... response, or the potential for the participant to respond with answers he or she believes the interviewer will perceive as favorable.
Panels
..... are longitudinal research designs. Data are collected through survey techniques at more than one point in time. In this case, a sample of persons is repeatedly measured over time. If the same individuals participate in each survey, or wave, then the panel is fixed. If some of the same individuals participate, and in addition, new participants are added, the panel design is rotating
Response rate
....., or return rate, is the number of people who respond after they have been contacted as part of the sample and asked to participate. An easy way to calculate .... is to divide the number of people who responded by the number of respondents identified as part of the sample.
Behavior
....... coding is the third type of pretesting and is used only when the survey is conducted in the face-to-face format. The idea in this type of pretesting is to have a third person monitor the interaction between the interviewer and respondent. The monitors are outside the survey process and, as a result, are more objective in identifying problems.
poll
A .... , like a survey, collects data only through a questionnaire, generally addresses issues of national or societal concern, and is conducted with a nationwide or statewide sample.
survey
A ...... is a system for collecting information. By asking questions or having participants respond to stimuli statements, researchers can collect data that can be used to describe, compare, or explain knowledge, attitudes, or behavior. Also known as a questionnaire or a poll, a ..... is probably the most commonly used (and perhaps misused) methodological tool for gathering information about almost any communication concept or topic.
expert panels
A fourth type of pretesting is the use of ...... In this type of pretesting, experts in research methodology or in the survey's content read through the questionnaire together and discuss potential problems they see with the survey. Because they are experts, this type of pretesting can point out semantic problems in how questions are worded and interviewer administration techniques, as well as potential problems with analyzing the data after the survey is complete.
conventional
A second type of pretesting is the ......pretest. In this type of pretest, the researcher selects several individuals who are like persons in the population. The survey is completed just as it will be done in the study. This type of pretest allows the researcher to reflect on the survey process and make any changes to the administration of the survey before it is given or sent to members of the sample.
semantic
Another common response set for closed questions is a ......differential scale. For each question or item, a respondent is given a numerical scale anchored by word opposites, or bipolar adjectives. -ex. friendly......unfriendly -satisfied....unsatisfied
pilot
As you can see, a researcher can spend consider- able time and energy either in looking for and adopting an existing questionnaire or in devel- oping a new questionnaire. This energy should not be wasted. Thus, it is a good idea to pretest the instrument. Pretesting, sometimes called .....testing, occurs when, before data collection actually begins, the researcher tries the survey or questionnaire with a small group of participants who are similar to those individuals who form the population. There are four approaches to pretesting a survey: cognitive, conventional, behavior coding, and expert panels
online
Because of their convenience for both researcher and participants, ..... or web surveys have become popular. They are also cost effective, do not have geographic restrictions, and can include color, graphics, and multimedia stimuli. Because digital media are used, survey responses are efficiently collected and managed in databases
scales
Communication Research Measures II: A Sourcebook, edited by R. B. Rubin, A. M. Rubin, E. E. Graham, E. M. Perse, and D. R. Seibold (2009), is a collection of ..... that measure communication constructs in the cross-cultural and intercultural, family, group, health, interpersonal, instructional, organizational, and mass communication contexts. Besides providing the scale itself, the book describes scale background, validity, and reliability, and also provides primary citations.
questionnaire
Survey is the appropriate term when a ...... is the only method of data collection, such as in a phone, face-to-face, or online survey
face
The most basic form of validity for survey questions is .... validity. A good way to establish face validity is to ask yourself the question "Does this item (or items) ask all the needed questions in language that is appropriate?"
self-administered
The most common survey form is one that is ...... in written form. In a .... survey, individuals read and select a response on their own. Often called self-reports, these can be done onsite, as in the research setting, mailed or e-mailed to individuals at work or home, or administered through online survey software. An advantage of self- reports for the participant is that the survey can be completed anonymously at the respondent's own pace.
Likert-type
The most common way to present closed questions and response options is the ......scale. This multi-item scale asks respondents to respond to survey items with these (or similar) choices: strongly agree, agree, neutral, disagree, strongly disagree. Technically, a single item is a Likert item; multiple items comprising a measure of a construct is a ..... scale
nominal
There are many different types of response sets for closed questions. One type is ....., or categorical. This means that responses in the set do not have a numerical equivalent. In a survey about communication choices at work, the researcher asks Are you (check the appropriate box): As explained in Chapter 5, .....or categorical responses are often used to obtain demographic information about respondents (for example, age, sex, marital status, employment status, socioeconomic status, education, income, occupation, religious preference or affiliation, race and ethnic background), and there are special considerations in developing these response sets.
nonresponse
Unfortunately, all surveys will suffer from ....., or the failure to obtain data from individuals in the sample. One other problem ex- ists: Some questionnaires or surveys are returned in an unusable form. In most of these cases, participants fail to fill out the survey in its entirety.
representative
First, the survey should be used as part of a sound research design. Second, the questions or items in the survey should be straightforward. Third, survey respondents, or the sample, should be chosen to be ...... of a population by procedures that are appropriate, given the research question or hypothesis. Fourth, a survey should be both reliable and valid. Fifth, participants' responses to survey questions should be analyzed within the context of the questions asked. Sixth, survey results should be reported accurately and ethically. This includes not using the data out of context.
random-digit
Generally, participants for phone polls are selected through ..... dialing (RDD). In this case, the phone number, not the person, is the sampling unit. Generally, researchers select a frame of telephone numbers, a large set of numbers in a particular area code that is coupled with a particular telephone exchange
response set
Options in a ..... should be exhaustive, mutually exclusive, and equivalent. Exhaustive means that all possible choices are represented. Mutually exclusive means that the respondent will view only one of the choices in the response set as the correct or best answer. The response set is not mutually exclusive if a respondent finds two choices that can represent his or her answer. Equivalent means that the response choices are equal to one another.
recall cue
In other open formats, participants are directed to recall episodes or past interactions they participated in. Researchers use a ...... to draw participants' attention to the issue they are interested in or to restrict their attention to a particular type of interaction. ..... that ask participants to begin with the most recent event are generally more effective unless the material has an inherent temporal or historical order—such as recalling where someone has lived since graduating from high school
cognitive
With .... pretesting, you would ask questions about the question "Do you own your own home?" When a participant answers "yes" to this question, does that mean personally owning the place where he or she lives, or does it mean owning the home with someone else? In answering "yes," does the person mean that he or she owns the home outright or that there is a mortgage on the home? In answering "yes," does the participant mean a single-family home or a condo, a high-rise apartment, or a duplex? -Thus, ..... pretesting allows the researcher to test for semantic problems, or problems affecting how easily the questions are understood.
construct
With respect to surveys, ..... validity is established when a survey in fact does distinguish between people who do and do not have certain characteristics.