chapter 6 Flashcards
What is the point of surveys and polls
a sample of people being asked to answer questions for research
Define an open ended question
A question that allows respondents to answer in anyway they see fit.
These questions provide spontaneous, rich info to observers
Down side the responses must be coded and categorized (difficult and time-consuming)
(Ex: “What are comments about this professor?”)
Define a forced-choice format
People give their opinion by picking the best of the two or more opinions
(Frequently used in political polls, EX: which might ask which of two or three candidates the respondent is most likely to vote for.)
Define Likert Scale
When a scale contains more than one item and is anchored by the terms strongly agree, agree, neither agree nor disagree, disagree, and strongly disagree
Define semantic differential format
A numerical scale used to rate a target object
EX: Rate my professor 1-5 rating for each category
What defines a well written question
- Clear and straight forward to answer
- The wording and order of the questions do not influence the respondents answers.
- Simple
Define double-barreled questions
A question that is two questions in one.
Has poor validity (because people might be responding to the first half of the question, second half, or both)
Define double negatives
Make the wording of survey questions unnecessarily complicated
The words impossible and never cause you to think in the double negative
Instead of measuring peoples beliefs it is more likely to measure their working memory or their motivation to pay attention
(Ex: “Does it seem possible to you that the nazi extermination of the jews never happened, or do you feel that it happened?”, “Guns should never be controlled)
Why does question order matter in a survey?
It can change the way respondents understand and answer the question later
(Ex: “How often do your children play” would have different meanings if the previous question was about sports vs music)
Define response set
a type of shortcut respondents can take when answering a survey.
When answering a set of related questions, some people might adopt a consistent way of answering all the questions that has little to do with their sincere opinions.
Tend to answer all the questions in the same way
Define acquiescence “yea-saying”
When people say “yes” or “strongly agree” to every item instead of thinking carefully about each one
Define nay-saying
When the respondent disagrees with every item
How can a researcher tell the difference between someone is yea-saying or nay-saying
By adding the reverse-worded items (reversing the question)
Define fence sitting
How can a researcher get rid of fence sitting
When respondents play it safe and answer in the middle of the scale
- Taking out the neutral answer on the scale (Ex: 3 on 1-5 scale)
- use a forced-choice format: pick one of the two answers
What are benefits and draw backs of self report?
Benefits: There’s some info. only you know about yourself,
Drawback: can sometimes be inacurrate, can be hard for people to explain why they are thinking, behaving, or feeling the way they do
Define socially desirable responding.
How can a researcher avoid these responses?
(sometimes faking good) These respondents know what they really think, but they are too embarrassed, shy, or worried about giving an unpopular opinion on the survey,
Researchers should ensure to their participants that their responses are anonymous, include “fillers” unrelated questions, use computerized measures to evaluate peoples opinions about sensitive topics.
Define observer bias
an observer records what they want to see or expect to see, rather than what is really happening
Define “masked” or “blind” study
a study design in which the observers are unaware of the conditions to which participants have been assigned, and sometimes are even unaware of what the study was about
Define observer effects (reactivity) and how to avoid it?
When people change their behavior when they know another person is watching
Avoid by:
- Making unobtrusive observations (hide, don’t be seen)
- Wait it out (let participants get use to the observers presence and forget they are there)
- Measure the behavior’s results (measure the trace of behavior left behind)
What’s a sample vs a population
Sample- Just part of the population
Population- The whole thing
When is a sample biased?
What causes a biased sample?
When the sample in unrepresentative of the population
Sampling only those who are easy to contact, sampling only those one is able to contact, sampling those who invite themselves, self selection (a sample is known to contain only people who volunteer to participate)
Define probability sampling
Probability sampling- drawing the sample at random from a population
Define simple random sampling
Simple random sampling- Most basic form of probability sampling (EX: Picking names out of a hat)
Define cluster sampling
Cluster sampling- Start with a cluster and then narrow it down randomly to a few (a sample)
Define multistage sampling
Multistage sampling- Start with a lot, narrow it down to a few, then take a sample from each group
Define random sampling
Stratified Random Sampling- Selects particular demographic categories on purpose then randomly selects individuals within each of the categories.
Define oversampling
Oversampling- A survey that includes an oversample adjusts the final results so that members in the oversampled group are weighed to their actual proportion in the population.
Define systematic sampling
systematic sampling- selecting two random numbers using a computer or random number table (Ex: say 4 and 7, Population full of students and the researcher would start with the fourth person in the room and then count off by choosing the seventh person until the sample was desired)
Define random assignment
Only used in experimental designs
When researchers want to assign participants into different groups (randomly) (often a treatment group and a comparison group)
Define convenience sampling
Samples are chosen merely on the basis of who is easy to access
Define purposive sampling
When researchers want to study only certain kinds of people, it makes sense to seek out those kinds of people
Is a bigger sample always a better sample?
Not necessarily.
