Process of assigning numbers or labels to attitudes, feelings, thoughts, objects, events and behaviors in accordance with specific rules for representing quantities or qualities.
a set of symbols or numbers so constructed that the symbols or numbers can be assigned by a rule to the individuals (or their behaviors or attitudes) to whom the blank is applied
Uses numbers to identify objects, individuals or groups for classification purposes. •Examples are gender (male, female), racial/ethnic group, child at home or not, bought or did not buy, attended or did not attend, etc. Are you.. ? Male or female.
Classification measurement AND supplies some information about the relative amount or gap between classification categories.
•Examples are rankings and ranking categories. •Please rank your top ten favorite movies.
Interval and Ratio scale
Scales that are ordered in a number line. Interval scales do not include zero. If zero is included, it is a ratio scale.
•Examples are 1 to 100 ratings scales, age, number of sales, number of time something occurs, etc.
The degree to which measures are free from random error and provide consistent data. A measurement scale that provides consistent results over time.
Repeat same measurement with different samples, approximating original conditions.
Equivalent Form Reliability
Question sets that measure opposite concepts are measured and will be correlated if reliable.
Internal Consistency Reliability
"Split-half" Sampling where respondents in same sample are randomly split and results are compared.
Split-half techniques: (in previous definition???)
The extent which measurement instrument captures characteristic being measured.
Judgment call. Does measurement "seem" to be measuring what it is supposed to measure?
Does measurement provide complete coverage of what is being measured, or is something missing?
refers to the extent to which a future level of a criterion variable can be predicted by a current measurement on a scale
a procedure for assigning numbers (or other symbols) to properties of an object in order to impart some numerical characteristics to the properties in question
Graphic Rating Scales
offers the respondents a graphic continuum typically anchored by two extremes.
Itemized Rating Scales
similar to graphic rating scales, except that respondents must select from a limited number of ordered categories rather than placing a checkmark on a continuous scale. Starting items are rotated on each questionnaire to eliminate the order bias that might arise from starting with the same item each time.
the respondent makes a judgment without reference to another object, concept, or person therefore itemized and graphic scales are. Asked about a single concept
the respondent is asked to compare two or more items and rank each item
Ex: Please rank order the following beginning with the one you like best
Baseball__ Basketball__ Football__ Soccer__
Respondents pick between two items and asked which one they want most. Items are randomly selected from a longer list of items. Items are scored and ranked starting with item selected most often.
Constant sum scales
Respondents distribute points across two or more items.
Semantic differential scales
Respondents rate different attributes of a concept, presented as bipolar opposites. Attribute ratings are averaged across all respondents to get an overall score for the concept.
A semantic differential scale, with a twist. It uses negative and positive numbers (usually -5 to +5), with the adjective in the middle. No need to come up with bipolar adjectives.
The most often used scale in marketing research. Individual statements are used to define concepts and respondents rate each statement across an ordered, labeled scale.
•Very quick and easy to construct, easy for respondents to understand and use, and easy to communicate in a report to clients and managers.
has the same number of positive and negative categories.
used when you know that attitudes are ordered more toward one side of a scale
A set of questions designed to generate the data necessary to accomplish the objectives of the research project.
the process of grouping and assigning numeric codes to responses.
those in which the respondents can reply in their own words
require the respondent to make a selection from a list of responses
the process of ascertaining that interviews were conducted as specified-goal of "blank" is solely to detect interviewer fraud or failure to follow key instructions
Involves checking for interviewer and respondent mistakes. The process of ascertaining that questionnaires were filled out properly and completely
One-way frequency table
the most basic tabulation. A table showing the number of respondents choosing each answer to a survey question
examination of the responses to one question relative to the responses to one or more other questions
The value below which 50% of the observations fall. Often used to summarize variables such as income or home prices when there is concern that there are extreme values that will distort the mean.
This is the value that occurs most frequently. There may be more than one mode in any data set.
This statistic measures the dispersion in a dataset. The sum of the squared deviations from the mean, divided by the number of observations minus 1, then squared.
the standard deviation squared, or the formula for standard deviation with the square root sign removed.
Another measure of dispersion, is equal to the maximum value for a particular variable minus the minimum value for that variable.
The researcher tries to determine (using a sample) whether some belief about a characteristic about a population is likely to be true.
Measurement of a variable in one population has no effect on the measurement of the other variable.
EX: Measurement of responses of men and women.
Measurement of a variable in one population may influence the measurement of the other variable.
Ex: Measurement of brand awareness at T1 and measurement at T2 among the same population. T1 measure affects T2.
-The exact probability of getting a computed test statistic that was largely due to chance
-The smaller the value, the smaller the probability that the observed result occurred by chance.
-The demanding level of statistical significance that can be met, based on the calculated value of the statistic
summarize categorical data
Are used to show measurements taken at defined particular points in time. Results may reveal an interesting pattern.
Plain Bar Chart
look up pic
Clustered Bar Chart
look up pic
Stacked Bar Chart
look up pic
-If a particular mathematical difference is large enough to be unlikely to have occurred by chance or sampling error
-refers to the degree to which one variable differs from another "statistically."
If numbers are not exactly the same, they are different (duh!). This does not mean that the differences are "meaningful!" (or statistically significant)
Managerially Important Difference
Many variables in a data set will be statistically different. The next question is "are the significant differences large enough to matter to management?"
The standard you are going to use to determine whether a difference is statistically significant or not
-Critical in the process of choosing between the null and alterative hypothesis.
not in notes but seems obvious
One-tailed vs two-tailed test
not in notes
Testing norminal/ordinal (categorical) scaled variables in a cross tabulationi
Testing Interval scaled variables
-Ex: Is the observed difference between a sample mean significantly different from what you would might expect by chance (or by sampling error)?
Analysis of Variance (ANOVA)
Testing Interval scaled variables and whether the means of the variables among subgroups are statistically different.