Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
53 Cards in this Set
- Front
- Back
Advantages of Surveys
|
1.) Respondents can answer large numbers of questions rapidly.
2.) Many people can be surveyed rapidly. 3.)You can make generalizations with a known level of confidence from your sample to a wider population. |
|
Disadvantages of Surveys
|
1.) Questions with a limited response options (yes/no) will give you numbers but no understanding as "why"
2.) Most survey designs do not allow us to access causal relationships. 3.) An increasing unwillingness of consumers to participate in them. 4.) Deciding whether an answer is valid or not. |
|
Cross Sectional Survey
|
typically a "slice of life" in that they capture what is going on at one point in time. (ex: a public opinion poll about a consumer product could have very different answers from one day to another).
|
|
Trend
|
measure the same items over time but draw different samples from the population each time.
|
|
Cohort
|
groups of people defined, most typically by having an event in common.
Example: all females in our class born in 1989; the graduating class of 2011 |
|
Panel
|
a group of individuals is sampled and recruited and the same individuals are retained to answer questions over time.
|
|
Question Writing
|
the 4 main questions that are asked on all surveys: sex, religion, marital status and age.
|
|
Open Ended
|
allows respondents to answer in their own words.
|
|
Dichotomous
|
force respondents to choose from 2 possible answers
|
|
Multiple Choice
|
provide several possible answers
|
|
Likert Scale
|
statements which respondents are asked to locate their level of agreement somewhere between "strongly agree" and "strongly disagree"
**Example: our in class surveys |
|
Problems with survey wording
|
Easy to misinterpret.
|
|
Double Barreled Questions
|
ask two questions simultaneously but only allow for one answer
|
|
Social Desirability Bias
|
??
|
|
Advantages of Phone Surveys
|
Most households have phones.
Can survey large samples in short time. |
|
Advantages of Mail Surveys
|
Give respondents time to consider questions and ability to answer in any order.
Good for delivering questions on complex issues that require thought. |
|
Advantages of Internet Surveys
|
Can be administered quickly
Can target special interest groups Can be emailed or posted to a Web Site. |
|
Snowball Sampling
|
Rely on members of a network to introduce you to more members of a different network.
ex: vegetarians. or vector recos |
|
Systematic Sampling
|
It's a system.
Ex: in a phone book, you might take every 10th person on the list |
|
Cluster Sampling
|
Going from big to small.
First sample larger units (states, provinces, counties) then sample towns and cities, then city blocks, then individual addresses |
|
Frequency
|
Number of times something occurs
|
|
Frequency Table
|
Everything compiled into a table.
|
|
Inferential Statistics
|
Helps us generalize about a wider population from a smaller sample of it.
|
|
Range
|
The maximum value minus the minimum value.
ex: range of scores on a test in class is between a 5% and a 75% so the range is 70% |
|
Variance
|
measure the extent of which scores in a data set vary.
|
|
Standard Deviation
|
The square root of variance
|
|
Disadvantages of Phone Surveys
|
Typically limited to a few short questions
Consumer resistance Caller ID's, etc. people don't answer |
|
Disadvantages of Mail Surveys
|
Low response rate (people might throw them out)
No way of knowing who completed the survey. Only targets people who are literate. |
|
Disadvantages of Internet Surveys
|
Might not be scientifically valid- only reflects the views of people who chose to respond.
Different browsers might view the survey differently (can't control the survey presentation) May not know who completed the survey May need email address, or phone number to drive respondents to web site. |
|
Population
|
Consists of every type of individual you want to study.
|
|
Population VS. Sampling Frame
|
??
|
|
Sample
|
a selected segment of a population presumed to represent that population. (Red, Blue, and Purple rectangles on powerpoint)
|
|
1936 Gallup vs. Literary Digest forecast illustrate about sampling?
|
They sampled only people who owned tv's and phones (?) which shows that they only surveyed wealthy people and that isn't a good random measure of a survey.
|
|
Census
|
study of every member of a population.
|
|
Census vs. Sampling?
|
??
|
|
Non-probability samples
|
based on a judgment by the researcher.
|
|
Convenience Sampling
|
based on convenience to the researcher.
|
|
Quota
|
one of the first survey designs that attempt to bring a scientific approach to survey research.
ex: the cop who needs to give people tickets to get his quota. so then think that we want to get 8 on-campus students and 2 off-campus students to take our survey. That is our quota. |
|
Probability Sampling
|
Involves random selection of the sample units.
|
|
Simple Random Sampling
|
throwing dice, drawing names out of a hat, and lotteries. There is no predicting what specific names or numbers will be sampled. (the luck of the draw)
|
|
Stratified Sampling
|
a way to "force" random groups into your sample.
(you set aside a certain % for a specific "group" such as vegetarians, etc.) |
|
Causal Relationship
|
a relationship between variables in which changes in one demonstrably result in changes in another.
|
|
3 Elements of Causality
|
1. A must precede B in time
2. If one thing changes, the other must change (covariance) 3. One variable must change the other, and not something else. |
|
Covariance
|
If one thing changes, the other must change.
|
|
Control
|
removes all other possible variables from the experimental design so we can be sure that our treatment variable and only our treatment variable is causing any change we see.
|
|
Control Group
|
Groups not exposed to any experimental variable.
|
|
Random Assignment
|
the use of random selection to assign research participants to experimental groups.
|
|
Mean
|
the average
|
|
Median
|
the middle
|
|
Mode
|
the number used most often
|
|
Measures of Central Tendency
|
the mean, median and mode
|
|
Confidence Level
|
the calculated probability of a value being true. **typically for communication research a confidence level of 95 is used, meaning that usually 95 out of 100 a reported value will occur if a population is repeatedly sampled.
|
|
Confidence Interval
|
a range of values, estimated from a sample, within which a value for a population is estimated to fall.
|