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;
82 Cards in this Set
- Front
- Back
Nominal
|
Purpose:
Classification Ex. Team player #'s |
|
Ordinal
|
Purpose:
Rank order Ex. 1st, 2nd, 3rd |
|
Interval
|
Purpose:
Measuring differences Arbitrary zero Ex. Temperature |
|
Ratio
|
Purpose:
Comparison of absolute magnitudes Zero is not arbitrary Ex. Profit |
|
Three properties of the numberying system
|
P1: Follow a specific order 9>3
P2: The difference between pairs of number can be compared Ex. 6-3=9-6 P3:We can divide one number by another and interpret the resulting ratio. Ex. 6 is twice as large as 3. |
|
Applicable number properties for Nominal
|
None
|
|
Applicable number properties for Ordinal
|
P1: Rank order
|
|
Applicable number properties for Interval
|
P1: Rank order
P2: Difference between pairs is measurable |
|
Applicable number properties for Ratio
|
P1: Rank order
P2: Difference between pairs is measurable P3: Two measurments can be compared. |
|
What are the central tendencies?
|
Mean: Average
Median: Middle value Mode: What occured the most |
|
What are applicable statistical tests for Nominal?
|
Chi-Squared
Percentages |
|
What are the applicable statistical tests for Ordinal?
|
Chi-Squared
Percentages |
|
What are the applicable statistical tests for Inverval?
|
Correlation Tests
ANOVA Regression |
|
What are the applicable statistical tests for Ratio?
|
Correlation Tests
ANOVA Regression |
|
Comparitive scales
|
A judgment comparing one object, concept, or person against another
|
|
Paired Comparison Scales
|
(Comparitive) Respondents are given objects two at a time and asked to select which one they prefer.
Number of comparisons: n(n-1)/2 |
|
Rank Order Scales
|
(Comparitive) Respondents compares one item with another and ranks them.
Problem: don't know amount by which one is preferred over another |
|
Constant Sum Scale
|
(Comparitive) Respondents are allocated a fixed number and asked to allot points based on relative importance totalling to that amount to each object.
|
|
Pictoral Scale
|
(Non-comparitive) Categories are shown graphically. Good for children and illiterate.
|
|
Continuous (Graphical scale)
|
(Non-comparitive) Rate objects by placing a mark in the appropriate position on a line running from one extreme to another.
Problem: hard to quanitify, use of decimals |
|
Likert Scale
|
(Non-comparitive) Respondents indicate thier own attitudes by checking how strongly they agree or disagree with statements.
|
|
Semantic Differential
|
(Non-comparitive) A series of seven-point rating scales with bi-polar adjuectives, such as good/bad, achore the ends. Weight is assigned to each position.
|
|
Stapel Scale
|
Use of a single adjective for the semantic differential when it is difficult to create pairs of bipolar adjectives.
-1-2-3<Wide selection>1 2 3 |
|
Image Profile
|
A graphical representation of semantic differential data for comparing brands, products or stores to highlight comparisons.
Stack semantic differentials and compare mean or median. |
|
Five issues that need to be addressed when designing a survey
|
1) Extent of category description
2) Number or response options 3) Treatment of respondent uncertainty or ignorance 4) Balance favorable and unfavorable categories 5) Strength of anchors |
|
Why is questionnaire design one of the most critical stages in survey and causal research?
|
It is one of the most critical stages because a survey is only as good as the design. Bad questions=bad results=bad decisions
Questions must be relevant and accurate |
|
What questions must be answered when designing a survey?
|
1) What should be asked? (Relevency)
2) How to phrase? (Accuracy) 3) What sequence? 4) What layout? 5) How to pretest? |
|
Open ended questions
|
Unstructured
|
|
Closed ended/ fixed-alternative questions
|
Structured:
Multiple choice Dichotomous Scales |
|
Seven mistakes of questionnaires
|
1) Complex language
2) Ambigous language 3) Leading questions 4) Double barreled questions 5) Making assumptions 6) Burdensome questions 7) Long questions |
|
What three criteria should you ensure that your response categories meet?
|
1) Responce categories are relevent
2) Responce categories are exhaustive 3) Responce categories are mutually exclusive |
|
How should you sequence questions?
|
Filter questions: screen out those who aren't qualified
Sensitive questions: put them toward the end Funnel: ask general before specific |
|
What overall order should questionnaires follow?
|
1) Qualify/ Screening
2) Introductory questions/warm ups 3) Main questions (easy) 4) Main questions (more difficult) 5) Psychographics/lifestyle 6) Demographics 7) Indentification info |
|
Pretesting
|
Testing a questionnaire on a small sample of respondents to identify and eliminate potential problems.
|
|
Experiment
|
A carefully controlled study in which the researcher manipulates a proposed cause and observes any corresponding change in proposed effect.
|
|
Experimental Variable (Indenpendent variable)
|
Respresents the proposed cause and is controlled by the researcher by manipulating it
|
|
Manipulation
|
The researcher alters the level of the variable in specific increments.
|
|
Dependent Variable
|
The effect (outcomes of interest) that are hypothesized to be influenced by the independent variable
|
|
Pre-experimental designs do not include:
|
Ransom assignment
|
|
What characteristic do True experimental designs have that pre experimental designs do not?
|
Such designs involve random assignment of participants to treatments.
|
|
Statistical designs
|
Such designs allow you to 1) examine the effects of different levels of an independent variable and 2) two or more indenpendent variables.
|
|
Types of pre-experimental designs
|
One-shot/After-only
One group Pre-test/Post-test Time series Control Group Design |
|
Types of True experimental desins
|
After-Only
Pre-test/post-test Solomon Four Group design |
|
Types of statistical designs
|
Completly Randomized Block
Randomized Block Design Factorial Design |
|
What are validity threats to One-shot/After Only Design?
|
History and Maturation
|
|
What are validity threats to One group pre-test post-test?
|
History and Maturation
|
|
What validity threat applys to all Pre-experimental designs?
|
Selection bias
|
|
How does time-series design address problems with pre-test/post-test design?
|
Addresses problems with history and maturation by taking several measures over time, both before and after period of interest.
|
|
How does control group design address problems with the other pre-experimental designs?
|
Address problems with history and maturation by including a control group. Helps identify any variables that may also be affecting dependent variable.
|
|
Internal validity
|
Reflects the accuracy of the measurement
|
|
What are the threats to internal validity?
|
History
Maturation Selection bias Mortality Testing |
|
History
|
Change in events between the beginning and end of the experiement that may influence the dependent variable
|
|
Maturation
|
Change in subjects during the course of study that effects the responce to the indenpendent variable
|
|
Selection bias
|
Not having randomly assigned groups
|
|
Solution to threats to internal validity
|
Use true experimental designs. There is less threat to internal validity but realism is sacrificed.
|
|
External validity
|
Reflects the generalizability of the experiment beyond the experimental situation. "Realism"
|
|
Threats to external validity
|
Artificiality of measures
|
|
Lab Experiements
|
Artificial settings
Isolates research under controlled conditions Short duration Less costly than field experiements Less external validity |
|
Field Experiments
|
Completely natural environment
Less control over conditions Long duration More expensive than lab experiments Greater external validity |
|
Sampling errors
|
Population errors
Non-responce errors |
|
Non-sampling errors
|
Ambiguity of questions
Interviewer error Ambiguity of answers Inaccuracy in response |
|
Population
|
Any complete group of entities that share a same common set of characteristics
|
|
Census
|
An investigation of all the individual elements that make up a population
|
|
Sample
|
A subset, or some part, of a larger population
|
|
Population element
|
An individual member of a population
|
|
Sample frame
|
A list of elements from which a sample may be drawn
|
|
Non-probability sampling methods
|
Convenience sampling
Snowball sampling Quota samling |
|
Probability sampling
|
Simple Random Sample
Systematic Sampling Proportional Stratified Sampling Cluster Sampling |
|
Convenience sampling
|
Obtaining those people or units that are most conveniently available to the researcher
|
|
Snowball sampling
|
Initial respondents are selected by probability methods and additional respondants are obtained from information provided by the initial responses.
|
|
Quota sampling
|
Ensures that various subgroups of a population will be represented on pertinent characteristics to the extent that the investigator desires
|
|
Simple Random sampling
|
Assures each element in the population has a known and equal chance of being included in the sample
|
|
Systematic Sampling
|
Order all units in the sampling frame and number them from 1 to N. Choose a random starting place from 1 to k and then sample every kth number after that. Let k=N/n
|
|
Proportional Stratified Sampling
|
The population is divided into subgroups (or strata) and then elements within each stratum are drawn in proportion ot the population size of that stratum.
|
|
Cluster Sampling
|
Population is divided in subgroups (or clusters) and clusters (not individual elements) are selected at random and all members of a subgroup are measured.
|
|
What one requirement is there for the chi-square statistic (hint: it concerns the expected frequencies)?
|
Expected frequencies must be at least 5
|
|
One Sample t-test
|
Used when comparing a sample mean to a specific value
|
|
Paired Sample t-test
|
Used when comparing means from one group on two different items
|
|
Independent Samples t-test
|
Used when comparing means from two groups on a single item
|
|
Hypothesis testing
|
The method used to prove or support arguments with statistics
|
|
Alternative hypothesis
|
What you are trying to prove, the new idea
|
|
Null hypothesis
|
The opposite of the alternative hypothesis
|