Quiz 2 Flashcards
(28 cards)
Basic details of the survey research design. What is a survey/poll? What does it allow us to do? What are the limitations?
- a method of posing questions to people on the phone, in personal interviews, on written questionnaires, or online
- psychologists might conduct national polls as part of their research, or they may use the polling information they read to inspire further research
- limitations: internal validity, missing data, unanswered questions
Types of survey questions (open-ended, forced-choice, rating scale). What are each? How do they work? What are the benefits and/or problems with each?
- open-ended questions introduce topic and allow participants to respond in their own words. they allow flexibility in choosing how to answer. however, not very specific.
- ex. What do you think about the current availability of food on this campus?, In your view, what are the most important factors in choosing a college or university?
- forced-choice questions involve people giving their opinion by picking the best of two or more options.
- benefit is they are more specific but only give a limited number of response alternatives
- ex. If the election were held today, which of the following candidates would receive your vote?
(a) Ms. Smith (b) Mr. Jones (c) Mr. Johnson
- rating scale questions require selecting a numerical value on a predetermined scale
Issues related to wording of questions (double-barrel, leading questions, negatively worded)
- double-barrel questions asks two questions in one. they have poor construct validity because people might be responding to the first half of the question, the second half, or both. Therefore, the item could be measuring the first construct, the second construct, or both.
- leading questions wording leads people to a particular response
- ex. “Opponents of voter ID laws argue the laws are intended to prevent certain groups from voting despite being eligible to vote. Do you strongly favor, mostly favor, mostly oppose, or strongly oppose voter ID laws?”
- negatively worded questions can cause confusion and mean that disagreement indicates a positive outlook on the topic
- ex. “Abortion should never be restricted.” or “It is difficult to get things changed in the agency.”
Observer bias and observer effects, what are each? How do they impact research? How can we prevent each?
- Observer bias is when instead of rating behaviors objectively, observers rate behaviors according to their own expectations or hypotheses.
- observer effects is when the observers unintentionally change the behavior of those they are observing, such that participant behavior changes to match observer expectations. This can happen to both humans and animals.
- To prevent observer bias, you use multiple observers and determine the inter-rater reliability of the observations
- To prevent observer effects: you can attempt to conceal the observer or try to habituate participants to the observer’s presence by repeated exposure until the observer’s presence is no longer something new/distracting
What is reactivity? Why does it happen? How can it be prevented?
- Reactivity is a change in behavior when participants know another person is watching. They might react by being on their best behavior, or their worst rather than displaying their typical behavior.
- it can be prevented by making yourself less noticeable, waiting out the situation until the participant(s) forget they’re being watched, or using unobtrusive data (for example, counting # of empty liquor bottles in residential garbage cans to see how much alcohol is being consumed in a community)
How might the order in which a series of questions are presented influence responses?
-earlier questions can change the way respondents understand and answer the later questions. for example, a question on a parenting survey such as “How often do your children play?” would have different meanings if the previous questions had been about sports versus music versus daily activities.
Behavioral observation. What information does it provide that surveys/polls do not?
- researcher observes and systematically records the behavior of individuals
- it provides information about what people are really doing or what really influences their behavior
- can sometimes be better than self-reporting, because people cannot always report on their behavior or past events accurately
Mechanics of a Likert scale
-this type of scale has more than one item and each response value is labeled with the specific terms “strongly agree”, “agree”, “neither agree nor disagree”, “disagree”, and “strongly disagree”
Populations vs Samples
-Population is the large group of people that are of interest for a research study
-Sample is the small group of individuals selected to participate in research study
•should be similar to population from which they are selected
•should represent the population of interest
What is a biased sample? What factors might lead the sample to be biased?
- a biased sample is also called an unrepresentative sample, in which some members of the population of interest have a much higher probability of being included in the sample compared to other members
- factors that might lead to this is when researchers study only those they can contact conveniently, or only those who volunteer to respond.
Difference between the target population and accessible population
o Target population is the entire set of individuals who have the characteristics required by the researcher
o Accessible population is group available to be studied
Generalizing results from samples to populations
the goal of research is to generalize results from the sample to the larger population
What is a census?
an official and complete count of the population in its entirety, including demographic information about members
snowball sampling technique
- you ask the participants who sign up to help recruit similar types of people
- this is if you are using a purposive sampling technique and are looking for specific type of participant
systematic sampling technique
-Every nth participant is selected from a list containing the total population
•starting position is chosen randomly
•example: start with person 6 and then select every 5th person
-Violates principle of independence but gives a high degree of representativeness
multistage sampling technique
- Starts the same as cluster sampling
- You randomly selecting preexisting clusters and then randomly sample from each cluster rather than testing every individual within the cluster
cluster sampling technique
- when you randomly select clusters (preexisting groups) from a list of all clusters that exist within the population
- Example: 10 classes of 30 students to obtain a total of 300 participants
- Benefits:
- Easy way to obtain a large, relatively random sample
- Measurement can be done in groups
- Problem:
- Selections are neither random nor independent
convenience sampling technique
-Participants selected based on those who are available and willing to participate
•Such as self-selection for an internet poll
-Easy method for obtaining a sample
•Our sample, psych 101 students, are convenience sampling
•Sample is probably biased
-It is a weak form of sampling, but it is extremely common
purposive sampling technique
-Restricting the sample to only certain types of people without using a random sampling technique
-Trying to have purpose to our selection
-Ex. New technique to treat nicotine addiction
•Post flyer at locations that cell tobacco products
-Is a convenience sample that only selects smokers
oversampling sampling technique
- the researcher intentionally over-represents one or more groups
- often used to make sure the number of participants is large enough to statistically analyze
simple random sampling technique
- participants are selected from a list of the total population
- ensures each individual has an equal chance and independent chance of selection
stratified sampling technique
- the population is divided into subgroups (or strata) and then randomly selecting equal numbers from each of the subgroups
- Ex. different income levels (families with incomes above $250,000, above $100,000 but below $250,000, ect.)
- Guarantees that each subgroup has adequate representation but the overall sample is usually not representative of the population
proportionate stratified sampling technique
-sample is obtained by subdividing the population into strata and then randomly selecting from each stratum a number of participants so that the proportions in the sample correspond to the proportions in the population
quota sampling technique
-subgroups are identified to be included (gender, age, etc.) and quotas are created to select individuals in each subgroup
•ex. rather than selecting the first 30 participants, regardless of gender, impose a quota of 15 females and 15 males. once the quota of 15 males is met, no other males can participate.
•allows for control over a convenience sample
-the problem is that sample probably is biased
