CHAPTER 4: Surveys and Interviews Flashcards
Use a way of obtaining information about people’s opinions, attitudes, preferences, and behaviors simply by asking.
- Allows us to gather data about experiences feelings thoughts, and
motives, that are hard to observe directly
Survey
questionnaires are tools, while interviews are the means of communicating the various items in the
questionnaire.
Questionnaire
is a research project all by itself, with steps that need to be followed to ensure that valid information is gathered and that the questions are really asking what the researcher intended to ask.
Constructing Survey
must be answered by one of a limited number of alternatives. This is the easiest to quantify.
(e.g. Do you smoke?, On a scale of 1-10 how much do you like Taylor Swift?)
Closed Questions (structured)
solicit opinions and feelings by asking the questions in such a way that the person must respond with more than a “yes or no”. It describes verbatimly.
(e.g. Why do you prefer detergent over liquid?) .
Open-ended Questions
quantifying answers to open questions must be designed to evaluate and categorize the content of each answer.
Content Analysis
ask responses about two (or more) different ideas in the same question which should be avoided.
(e.g. Do you like strawberries and cream? (vice versa) that prevents you from answering the question.
Double-Barreled Question
they contain emotion-laden words, being radical or racist should be avoided.
(e.g. What can you say about the “Black people ‘’ terror attack on the “whites’ in Wisconsin?, Most people favor the use of reproductive pills. What do you think?)
- To avoid bias, it is better to mention all possible alternatives or to mention none.
Leading or Loaded Questions
starting from the most general questions and moving towards the most specific questions pertaining to a given topic. This is also good in qualitative research.
Funnel Questions
general questions asked to respondents to find out whether they need to be asked more specific questions.
- If the answer is NO, the interviewer should not ask the specific questions.
Filter Questions
creating open-ended questions requires the subject to answer in a narrative form rather than selective from the choice or responses.
Measuring Responses
scale used to measure responses.
Level of Measurement
classifies response items into two or more distinct categories on the basis of some common features. (lowest level of measurement because it provides no info about magnitude).
- Either answer “True or False” but not both
Nominal Scale
rank ordering of response items. (Ranking of candidates of presidential polls, beauty contests, most popular , 2nd, 3rd and so forth.
Ordinal Scale
measures magnitude or quantitative size using measures of equal interval between values.
- No true zero point. Has intervals can go to negatives: -2,-1, 0, 1,2
Interval Scale
the highest level of measurement, and it has equal intervals between all values and a true zero point. Physical properties such as height, weight are characteristics whose quantity or magnitude can be measured using ratio scales.
Ratio Scale
are more powerful and preferred by researchers than those of nominal and ordinal.
Interval and Ratio data
may be regarded as traits, attitudes, and preferences that any individual could fall at any point along each dimension.
Continuous Dimension
subject do not have to think about a great deal (no open-ended questions)
Relevant to the Central topic
subjects are able to answer without saying “i don’t know”.
Easy to answer
subjects will think are relevant to the topic of the survey.
Interesting
tendency for subjects to respond to questions or test items in a specific way, regardless of the content.
Response Style
comes into play whenever questions require specific knowledge about facts or issues. When unsure, some will leave it blank or make a wild guess. Unwillingness to answer is the most often problem.
Willingness to answer
in multiple choice exam response style can influence the answer, when in doubt sometimes fall into the middle.
Position Preference
the plain meaning of the words that actually appear in the page.
Manifest Content
apt to agree with a question regardless of its manifest content. (Answer “YES or TRUE” items to most items)
Yea-sayers
tend to disagree no matter what they are asked. (Tend to say “NO or FALSE” to most items).
Nay-sayers
the position of the question - when it falls within the question order can influence how the question is interpreted.
Context Effect
Sometimes it helps to separate related items and questions unrelated to either of the related questions.
Buffer items
description of relatively permanent aspects of subjects or settings. (E.g. Determining the signs of homosexuality: SPO)
Static Checklist
records the presence or absence of specific behavior. (E.g. A classroom observer may note whether children are talking or indeed quiet.)
Action Checklist
permits only define responses.
Closed Form
allows respondents to make freewheeling responses.
Open Form
present open-ended questions that allow respondents in any way they wish. Has a high probability that the respondents will think more or authentic answer rather than choosing from a set of responses predetermined by the researcher.
Open-ended Format
presents questions followed by a set of options predetermined by the researcher based on a pre-survey. The format is extremely structured, easier to get responses and analyze data. Appropriate in situations where researchers know the full range of possible responses to their questions.
Multiple Choice
is a question that can have two possible answers. Questions are usually used in a survey that asks for a Yes/No, True/False, Fair/Unfair or Agree/Disagree answers.
Dichotomous
a questionnaire using a totally structured format and occasionally with free responses option, the difference between multiple choice and checklist form is that respondents can answer a single question with one or more options.
Checklist Format
Measures:
- Frequency (how many times)
- Duration (how long)
- Rating (majority/minority)
a purposeful face-to-face relationship between two persons: the interviewer who asks questions to gather information and the interviewee or respondents who supplies the information request.
Interview
“the hidden meaning behind the question.”
Latent Content
a picture we want to create of ourselves. Trying to put our best foot forward (distorting the truth).
Response Set
- Be sure to include a polite and professional cover letter (stamped, self-addressed envelope)
- Return procedures protect the subject’s anonymity.
- Report unreturned questionnaires.
Mail Survey
- Having access to computers, consider a computer-based survey.
- Web - based if you have internet programming skills.
- Data should be kept confidential.
Computer and Internet Survey
- Must establish rapport quickly with strangers.
- Needs well-versed/trained interviews who can obtain high-quality answers.
Telephone Survey
need to spend time to establish trust and win the subject’s confidence. Factors such as (good voice/Halo effect; accent; SES; Race).
Face-to-Face
the same questions are asked in precisely the same way each time (unstable; quantifiable). (as is)
Structured
are more free flowing, free to explore interesting issues as they come up, but the information may not be usable for content analysis or stats.
Unstructured
usually small groups of people with similar characteristics (all foreign students; members of SHE; faculty members of a certain dept. ) brought together by an interviewer called “facilitator”.
Focus Groups
the extent to which the survey is consistent and repeatable.
Reliability
refers to the extent to which a survey usually measures the intended topic.
Validity
scientific observation is made under precisely defined conditions, in a systematic and objective manner, and with careful recording keeping.
Observation
covers the entire population.
Universal
covers a representative of a population.
Sample Survey
deciding and selection of potential subjects.
Sample Technique
consist of all people, animals or objects that have at least one characteristic in common. (E.g. LGBTQ+ , IAS, Teume)
Population
subset population of interest (e.g. either lesbian, gay, bisexual, queer, transexual, intersexual, asexual, allies)
Sample
study the likelihood of events.
Probability
closely mirrors the larger population (e.g. population - IAS - Mass Com. , Psych, MedTech)
Representative
selecting subjects in such a way the odds/chances of being in the study is being known or can be calculated.
Probability Sampling
unbiased method of selecting subjects such as flipping a coin, drawing a number from a hat – (fish bowl technique) each member of the population has an equal opportunity of being selected, and the outcomes cannot be predicted ahead of time by any known law.
Random Selection
the most basic form of probability sampling whereby a portion of the whole population is selected in an unbiased way. Does not guarantee that your sample truly represents the whole population. It yields an estimate of what is likely to be true.
Simple Random Sampling
elegant variation of random sampling where all members of population are known and can be listed in an unbiased way.
Systematic Random Sampling
This type of analysis is of value only when we measure “traits”
or characteristics that do not change over time.
- IQ and stable traits.
The Test-Retest Method
This effect occurs when the first testing session
influences scores from the second session.
- This effect occurs when the first testing session
influences scores from the second session.
Carryover effect
The order of administration is usually counterbalanced to avoid practice effects.
Parallel Forms Method
a test is given and divided into halves that are scored separately. The results of one half of the test are then compared with the
results of the other.
- If the items get progressively more difficult, then you might be
better advised to use the odd-even system
Split - Half Method
the internal consistency of a test refers to the intercorrelations
among items within the same test.
Factor Analysis
These reliability estimates have various names, including interrater,
interscorer, interobserver, or inter judge reliability
Inter - Rater Reliability
is the mere appearance that a measure has validity.
- can motivate test-takers
Face Validity
of a test or measure
considers the adequacy of representation of the conceptual
domain the test is designed to cover.
- of a test or measure
considers the adequacy of representation of the conceptual
domain the test is designed to cover.
Content-related evidence for validity
describes the failure to capture important components of a construct.
Construct Underrepresentation
occurs when scores are influenced by factors irrelevant to the construct
Construct - Irrelevant Variance
tells us just how well a test corresponds
with a particular criterion.
- For example, a test might predict success in marriage. Marital success is the criterion, but it cannot be known at the time the
couples take the premarital test
Criterion validity evidence
assessments of the simultaneous
relationship between the test and the criterion.
- For example, if you want to know who will perform well among applicants, look at the criterion set by successful employee.
Concurrent Validity Evidence
we can have reliability without validity. However, it is logically impossible to demonstrate that an
unreliable test is valid
Researchers sample entire cluster – whole group is selected rather than individuals.
- For example:
Countries
Philippines, Thailand, Vietnam
Every one from the cluster can be surveyed
Cluster sampling
selecting samples through predetermined quotas that are intended to reflect the makeup of the population. They can reflect the proportions of important population subgroups, but the particular individuals are not selected at random.
Quota Sampling
a convenience sample is obtained by using any groups who happen to be convenient; considered a weak form of sampling because the researcher excercises non-control over the representative of the sample.
Convenience Sampling
non random samples are selected because the individuals reflect a specific purpose of the study. Usually sets up a criterion/criteria on the target research that may be useful to the study.
Purposive Sampling
sampling procedures in which subjects are not chosen at random.
Non Probability Sampling
a non random sample wherein people who fit the sample criterion and asks these people to locate them to additional individuals.
Snowball/Network Sampling
the intervals in which observations are made; chosen either systematically or randomly with a representative sample of behavior.
Time Sampling
efficient method of behavior sampling when behavior occurs infrequently.
Event Sampling
behavior should be observed under many circumstances; location or condition.
Situation Sampling
the process of abstracting and summarizing behavioral data.
Data Reduction
the degree to which two independent observers are in agreement.
Inter - Observer Reliability
the proportion or percentage of times that various behavior occur.
Relative Frequency
participant wants to cooperate and what behaviors are expected with them to be a “ good participants”.
Demand Characteristics
