Exam 4 - Chapter 13-17 Mrkt Research Flashcards
Continuous Variables?
Large number of choices, not infinite but close
Validity?
Independent variable causes the dependent variable to happen
Reliability?
Test or survey is reproducible over and over again, may be wrong answer but you keep getting same answer
Sensitivity?
The scale is appropriate to the variable being measured, not high
Index Scales?
One that are put with number aptitudes
Concurrent Validity?
Two things happen at the same point in time, and one of them seems to give them the same kind of answer that the other does
Face Validity?
Where experts look at your questions and tell you this is going to be a valid survey
Predictive Validity?
Predicts something in the future. Example of ACT or SAT score predicting someone’s life or career
Affective Attitudes?
Emotional
Cognitive Attitude?
Belief
Behavioral Attitude?
Intention to do something or not to do something
Conceptual Definition?
Words or phrases that define descriptively what is to be measured
Operational Definition?
A recipe indicating specifically how to calculate a measure if some concept or construct
Nominal Scale?
One that simply identifies
Ordinal Scale?
One that shows thing in order
Interval Scale?
One in which the differences in the scale itself are the same like the temperature scale, the time
Ratio Scale?
Has a zero point
Attitude Characteristics?
Hypothetical construct, can not measure attitude directly. Can only measure by what people say or do
Randomized Response Questions?
Used in a situation in where you are not doing a paper survey but having a direct conversation with someone. Its a sensitive question so you give them a choice of two questions that they can answer and they only know what they answered
Graphic Scale?
Straight line with a bunch of ticks on it, can be smiley faces things like that
Likert Scale?
Declarative sentence followed by disagree, somewhat disagree, neither of these, agree
Staple Scale?
Plus or minus, such as friendliness. +3, -5
Semantical Differential Scale?
Seven point bipolar scale which has a descriptor at each end, which you may come to be trustworthy or not trustworthy
Constant Sum Scale?
Limited to three or four totals. Percentages as to reasons for doing something, what influenced you? adds up to 100%
Checklist Scale?
One in which you give people a whole list of answers, what magazines do you read, what TV shows you watch? Let them pick a bunch of different answers. Often used in pre-screening people who may participate in another survey
Ranking Scale?
Put things in order but there is no number on them, you can have ties. One has to be better than others, no zero point. People could not like any of them
Rating Scale?
You can have ties, research more
Sorting Scale?
Put them in kinds, different color schemes, whatever
Even Scale?
No center point
Odd Scale?
Center point
Balanced Scale?
Same number of positive responses as their are negative responses,
Unbalanced Scale?
Skewed to one side or the other. More good or bad answers, do this when you know people are going to be inclined to one side or the other
Forced Scale?
One that does not have a knockout, “no opinion”“Do not know”
Unforced Scale?
Has knockout, no option is
Questionnaire Examples and problem areas?
They don’t remember and they’ll tell you they do remember because they are embarrassed about not remembering. So they’ll write something down…
Calculation Problem Areas?
Force them to make a calculation they don’t know
Double-barreled Question?
Two questions in one
Leading or Loaded Question?
Which means it suggests the answer desired
Ambiguous Question?
Not clear what is being asked
Complex Question?
Too complicated to understand the question
Use of Jargon Words?
Using words that are not in people’s vocabulary, usually aim these at 6th graders
Collectively Exhaustive?
All of possible responses are covered, their isn’t a response someone is looking for that isn’t there
Mutually Exclusive?
No overlap, can not choose two things. No two players on the football team can have the same number
Counter-biasing Statement?
You try to reduce the potential embarrassment people may have, “its common in the US to have this sort of situation, given that what is your opinion”.
Potential for biasing un-intenionally by doing that. Do that only when you think their is a high likelihood when people will be embarrassed.
Pivot Question?
Move’s you to different points of the survey
Filter Question?
Qualifies you to take the survey or not take the survey
Survey Sequence?
From general questions to specific questions, “income, age” Want people to almost complete a survey before asking questions like this. (Classification questions)
Probability Sampling?
A situation in which, the probability of drawing any individual sample is known.
Non-Probability Sampling Techniques?
Chance of drawing an element not known. Using quota samples, here’s some people I want you to pick out in the mall
Characteristics of types or Probability Sampling Techniques?
Simple Random(Each element in the sample frame has an equal chance of being drawn),
Stratified(sampling each of several homogeneous subgroups separately),
Cluster(Sampling elements in a heterogeneous grouping of sampling elements where the groups to be sampled are draw at random from among the total number of such groups,
Area(Don’t need the names, just geography)
Characteristics of types or Non-Probability Sampling Techniques?
Convenience(people available in a class, church group, on a street),
Quota(specifying the number of respondents in several categories and subcategories),
Judgement(Samples chosen by experts…. Election exit polls),
Snowballing(Assembling list by referral, for widely-dispersed or rare groups)
Central Limit Theorem?
Talks about the distribution of means in a statistical array, the means will be normal as long as one of two things is true: The underlying distribution is normal. The sample size is greater than 30
Measures of Central Tendency?
Average(mean), Middle point(median), Most frequently occurring one(Mode)
Distribution Spread?
Standard Deviation, Range(end to end distance 4-20,16 is range), Variance
Confidence Interval?
The mean that you choose is within a certain distance of the truth?…. A specified range of numbers within which a population mean is expected to lie
Discrete Variables?
Variables that can only take on a finite number of values
Dichotomous Variables?
gender: male/female, over/under 65
Concepts?
A generalized idea that represents something of meaning.
Concrete Concepts: Age, Sex, Income
Abstract Concepts: Brand loyalty, corporate culture
Construct?
Abstract concepts measured by multiple variables
Inferential Statistics?
With inferential statistics, you are trying to reach conclusions that extend beyond the immediate data alone. For instance, we use inferential statistics to try to infer from the sample data what the population might think.
Descriptive Statistics?
We use descriptive statistics simply to describe what’s going on in our data.
