Term
|
Definition
A study using questionaires or interviews to discover sescriptive characteristics of phenomena |
|
|
Term
|
Definition
Follow-up interview questioning that ask respondents to explain how they feel about the interview |
|
|
Term
|
Definition
A questioning strategy that starts with an open-ended question and follows up with increasingly narrow questions |
|
|
Term
|
Definition
A questioning strategy that starts with specific questions and expands to more general questions |
|
|
Term
|
Definition
A strategy of follow-up interview questioning that directly asks for elaborating and explaining |
|
|
Term
|
Definition
A type of study using experimental methods in which researchers examine the effects of variables manipulated by the researchers in situations where all other influences are held constant. Variables are manipulated or introduced by the experiementers for the purpose of establishing causal relationships |
|
|
Term
Randomization (principle of randomization) |
|
Definition
Assignment so that each event is equally likely to belong to any experiemental or control condition. |
|
|
Term
|
Definition
Selection of data such that each event in the population has an equal chance of being selected |
|
|
Term
Selection (selection bias) |
|
Definition
A source of internal invalidity involving sampling biases in selecting or assigning participants to experiemental of control conditions (in essence, rigging the study by taking samples capriciously) |
|
|
Term
|
Definition
A source of internal invalidity involving changes that naturally occur over time (including fatigue or suspicion), even if subjects are left alone. |
|
|
Term
|
Definition
A source of internal validity involving biases introduced when subjects differentially (non-randomly) drop out of the experiment. |
|
|
Term
|
Definition
A source of internal invalidity involving changes in the use of measuring instruments from the pretest to the posttest, including changes in raters or interviewers who collect the data in different conditions |
|
|
Term
|
Definition
A source of internal invalidity in which events not controlled by the researcher occur during the experiment between any pretest and posttest. |
|
|
Term
|
Definition
A source of internal invalidity involving alternations that occur when subjects are tested and made testwise or anxious in ways that affect them when they are given a second test. |
|
|
Term
|
Definition
The universe of events from which the sample is drawn |
|
|
Term
|
Definition
The degree to which a sample differs from the population on some measure |
|
|
Term
|
Definition
Selecting events from a population |
|
|
Term
|
Definition
A range of values of a sample statistic that is likely, (at a given level of probability, called a confidence level) to contain a population parameter |
|
|
Term
Critical Value (critical alpha) |
|
Definition
In statistical significance testing, the line that divides the critical region from the rest of the probability distribution. |
|
|
Term
|
Definition
Though computed differently, a measure that attempts to summarize the average of squared differences of scores from the mean, symbolized from the sample variance as s2 and for the population variance as o2 |
|
|
Term
Standard Deviation (definition) |
|
Definition
Though computed differently, a measure that attempts to summarize the average deviation of scores from the mean, by estimating such a value from the square root of the variance of s2 symbolized for the sample standard deviation as s. |
|
|
Term
|
Definition
A family of nonparametric tests that permit examining observed frequencies of events with expected frequency. |
|
|
Term
|
Definition
A test of statistical significance designed to assess the difference between the means of two groups |
|
|
Term
Two-tailed Hypothesis (nondirectional material hypothesis) |
|
Definition
Hypotheses that state simply that there will be some kind of relationship between variables |
|
|
Term
3 Evidence for making a convincing causal argument (a reason that a given factor is responsible for producing certain other results) |
|
Definition
1. Association (co-variation)- when a presumed cause changes the effect has to change too. 2. Direction of Influence- time order 3. Control- the elimination of rival explainations for an association. The association must remain when other potential causes are eliminated. |
|
|
Term
Structured, Semi-Structured and Unstructured survey |
|
Definition
Structured: use a specific list of questions Semi-Structured: Some questions set in advance, some response categories set in advance, active probing and improvising base on responses Unstructured: interviews that permit respondents to indicate their reactions to general issues without guidance from highly detailed questions |
|
|
Term
3 Ways to Increase or Decrease Reliability |
|
Definition
1. Check questions-using the same question twice in different parts of the survey, one negative and one positive 2. Test-taking measures-uses random questions to help establish a person as lying, responding in a socially desirable way, or answering randomly without reading questions 3. Polarity rotation-alternating positive and negative adjective in questions, avoid condensing all positive in one portion of the quiz |
|
|
Term
Random Assignment vs. Random Sampling |
|
Definition
Random assignment: to assign so that each event is equally likely to belong to any experimental or control condition. Random sampling: selection so that each event in the population has an equal chance of being selected. |
|
|
Term
3 Reasons for Network Analysis |
|
Definition
1. To construct a map of the interaction 2. To diagnose problems in communication flow 3. To identify roles played by different group members |
|
|
Term
Surveys are most useful when people: |
|
Definition
1.are likely to have generally high or low sensitivity to the questions. 2.can be placed into just a few, easily understood categories. 3. questions and answers can be formatted in a standardized way. 4. When you want a large sample but have few resources |
|
|
Term
Surveys are not the best choice when: |
|
Definition
1. The sensitivity of respondents is highly variable or unknown. 2. When you have very complex questions. 3. When answers are difficult, not likely to be recalled accurately. 4. When you do not have a complete set of questions, but want to improvise as you explore subjects’ responses. 5. When you have a lot of different types of questions. |
|
|
Term
Steps to Designing a Survey |
|
Definition
Step 1: Decide exactly what to measure Step 2: Draft & revise questions systematically. Step 3: Take the “final” questionnaire for a “test drive” with subjects. Step 4: Think through administrative issues. Step 5: Do final revision based on what you learned in Steps 3 & 4. Step 6: Administer your survey to real subjects |
|
|
Term
Single most important step when designing a survey? |
|
Definition
|
|
Term
Interviews are a good idea when: |
|
Definition
1. The questions you are asking are not highly sensitive 2. Responses are less likely to fall into categories you know in advance. 3. You need multiple types of questions and formats. 4. You want to note the nonverbal that come with a verbal response. 5. You need to explain the context in order for respondents to understand what you’re getting at. 6. You want to probe their answers to gain a deeper understanding. 7. When it's difficult to get people to pay attention to a survey form |
|
|
Term
It is a poor or overly costly choice to conduct an interview when: |
|
Definition
1. The questions are simple and respondent sensitivity is low. 2. When you know how most of the answers will be categorized before 3. When answers are all easily recalled. 4. When all or most of the questions can be formatted the same way. |
|
|
Term
Deciding how structured to make a survey |
|
Definition
1. More when you have clear ideas about what you want to ask and the types of answers that are most likely. 2.More when time and costs are a concern-less structured interviews take longer and are more expensive. 3. Less when you want to gain depth about each respondent’s unique experience and perspective. 4. More when you want to compare responses more easily or systematically 5. Less to make participants feel personally consulted and involved 6. More when you want responses from a large number of people |
|
|
Term
3 Ways to Increase Readability of Research Material |
|
Definition
1. Shorter sentences 2. Smaller words (fewer syllables) 3. Use active sentences rather than passive ex: subject-verb-object |
|
|
Term
3 Ways Network Analysis Helps us Understand Stuff |
|
Definition
1. Problems that may disrupt the intended flow of communication 2. The roles that people play in groups 3. Structures in the pattern of communication among individuals or groups. Allows you to map the interaction. |
|
|
Term
Of the three types of research strategies: (surveys, interviews and experiments) which gives the best evidence of causal reasoning when done well? Why? |
|
Definition
EXPERIMENTS!! *b/c you can isolate variables, & obtain the clearest evidence of co variation *b/c you control the timing of when the independent variables appear, you have the best possible evidence of time-order. *b/c you control so many facets of the study yourself, you have more opportunities to rule out rival hypotheses. To eliminate “nuisance variables” – sources of rival hypotheses. |
|
|
Term
Six strategies to rule out rival hypothesis: |
|
Definition
1.Elimination and removal—by using a laboratory or controlling the setting, researchers can simply eliminate many factors that might influence results. This lets them focus on the factors of greatest interest. 2. Holding constant—by holding one variable constant, you can get a better picture of how two other variables might be related. 3. Matching—pairing subjects in groups so that the same kind of people show up in each of the groups you are trying to compare. (ex: having equal men and woman in both groups if we know woman brush more often than men. 4.Blocking—build the source of the rival hypothesis ( “nuisance variable” ) into the study where you can measure it directly. 5.Randomization—balance the groups you’re comparing by placing subjects into groups randomly. Take advantage of the principle of randomization. 6. Statistical Control—if you measure a “nuisance variable”, the source of a rival hypothesis, you can use statistical tools to turn it into a constant mathematically. |
|
|
Term
How does “pre-test-posttest control group design” help rule out rival hypotheses? |
|
Definition
Definition: a true experimental design that includes randomly selected experimental and control groups, each of which is given a pretest and protest |
|
|
Term
Difference between random sampling and random assignment |
|
Definition
-Random sampling deals with how you select subjects from the population. -Random assignment deals with -how you place them into groups after you have selected them. Also called randomization. |
|
|
Term
|
Definition
Convenience sampling- selection of events that are most readily available Quota sampling-samples are defined based on the known proportions within the population, and nonrandom sampling is completed within each group Systematic sampling- a method by which researchers select respondent according to a predetermined schedule rather than a random sequence Simple random sampling- a method by which researchers select participants or events such that each event in the population has an equal chance of being selected Stratified random sampling- a method by which researchers select participants or events to represent known proportions of characteristics in the population. After population characteristics are identified, (such as the number of men and women in the population) a random sample of a given size is drawn from each population stratification variable consistent with the population proportions. |
|
|
Term
What is central tendency of distribution and what are three ways of measuring it |
|
Definition
Central tendency of a distribution are measures that report averages of different varieties 1. Mean 2. Median 3. Mode |
|
|
Term
Why is the median better than the mean at measuring central tendancy? |
|
Definition
The median is a better measure of central tendency than the mean when there is an outlier than may affect the average dramatically. |
|
|
Term
What is dispersion or variability? |
|
Definition
Dispersion or variability is different ways to measure variability: 1. Range 2.Variance 3. Standard deviation 4. Standard error |
|
|
Term
Why is the range a bad measure of variability? |
|
Definition
The range is a poor measure of variability because it doesn’t tell you anything about the numbers between the highest and the lowest |
|
|
Term
What is the standard error ? |
|
Definition
The standard error is an estimate of how far from the population mean (true mean) the mean of a given sample is like to be. The variation and sample size effect how large it is. |
|
|
Term
What does it mean to say that a test result is statisticially significant? |
|
Definition
If a test is “statistically significant” it has an association that is beyond what is unlikely to occur by chance alone. |
|
|
Term
| If a pattern in a sample has a very low likelihood of occurring by chance, (say less than 5%), what can we say about whether they are likely to occur in the population? |
|
Definition
It should also be true for the population |
|
|
Term
Rival Hypothesis vs. Null Hypothesis ex: people who wear pink are usually younger |
|
Definition
Rival Hypothesis: people who wear pink tend to be older Null Hypothesis: there is no correlation between age and people who wear pink |
|
|
Term
|
Definition
As the independent variable changes the dependent variable also changes |
|
|
