Final Exam Business Research
Research process activities
Collect the data Enter the data Prepare the data: post coding data and editing data Examine the data
Appendices
Complex tables Statistical tests Supporting documents (research request for proposal and research contract) Copies of interview forms, observation checklists, measurement instruments Instruction to data collectors
Objects
Concepts defined by ordinary expensive
Management dependent variables
Conflict resolution Employee satisfaction Employee turnover
Construct validity
Consider both theory and measuring instrument being used
Classification questions
Cover demographic, economic, geographic, and sociological variables Can use screening and filter questions
independent; dependent
In regression analysis, the symbol X is commonly used for the ______ variable, and the symbol Y is commonly used for the ______ variable.
Metric
Interval Ratio
Oral and Written reports
Introduction Background Methodology Findings, insights and recommendations
Non-question elements assist the participant to answer the questions in the survey
Introduction Conclusion Transition Instruction
Findings, Insights, and Recommendations
Largest portion of report Determine the order of findings and insights Recommendations used for managerial action Include suggestions for alternatives supported by data and insights May recommend further initiatives
Actionable insights
Lead to recommendations Align with business goals and strategies Clearly communicated
Role of Researcher in Reporting
Make sure audience understands how insight was reached Provide correct and sufficient support Helps audience understand its value and embrace the insights
Marketing independent variables
Marketing mix: Price, product, promotion, distribution
Semantic Differential Scale
Measures the psychological meanings of an attitude object and produces interval data
Regression Analysis
Measuring the linear association between a dependent and an independent variable Regression is a dependent technique Makes a distinction between dependent and independent variables Specifies cause and effect relationships
Median
Midpoint of a distribution
0.05
Most common significance level in research is ______________.
Mode
Most frequently occurring value
r=0
No correlation
Nonmetric
Nominal Ordinal
nonmetric; metric
Nominal and ordinal scales are examples of ______ scales, while interval and ratio scales are examples of ______ scales.
Nominal
Numbers written on race cars, numbers assigned to object serve as tags to categorize or arrange objects in class
5 Factors of Unstructured/Structured questions
Objectives of the study Participants level of information about the topic Degree to which participant has thought through the topic Ease with which participant communities Participants motivation level to share information
Order Topics and Questions
Order topics based on participant interest, not research objective or investigative questions Order topics to establish participant rapport interest, and maintain engagement Cluster related topics together Order classification questions (not used as screens or filters) at the end of the instrument
Administrative questions identify
Participant Interviewer Interview location Research conditions
be easy raters
Participants tend to ________ when they make an error of leniency.
Introducrtion
Prepare audience for report Establishes researcher's credibility Contains: Title, researchers' profiles (name, title, company, educational, and experiential background), Executive Summary, and Table of Contents
Structured
Present the participants with a fixed set of choices; closed questions
marketing independent variable
Price, product, promotion and distribution can be categorized as ______________________.
Unstructured
Provide the participant with a frame of reference for their self-crafted answer; open-ended questions
Background
Provides audience with the following: Knowledge critical to interpreting the research findings Establish credibility Establish validity of the findings
Interval
Quantify the difference between values-temperature scale because difference between each value is the same- no true 0
Crafting Measurement Questions
Question coverage Question wording Question's frame of reference Question personalization Structure, number, or order of response alternatives
Data
Raw Unprocessed Collected quantitatively and qualitatively
Criterion-related validity
Reflects the success of measure used for prediction or estimation
Predict y using x values
Regression analysis seeks to ________.
Hypothesis testing
Relational hypotheses Differences of groups hypotheses Differences of standard hypotheses
Compiled Content and Style Infographs
Relatively new concept for written reports Blends data with design in picture format Receives more information visually then other senses Prime the audience and stimulates interest Makes information easy to understand Retention of data presented Visual elements can be utilized to represent relations between different types of data
Selecting a Measurement Scale
Research objectives Response types Number of dimensions Balanced or unbalanced Forced or unforced choices Number of scale points Rate errors
Significance levels and P-values
Researchers specify an acceptable significance level Most common level is .05 If p value resulting from statistical test is less than .05, hypothesis about difference is supported As the observed value or observed mean gets further away from standard mean, p value gets smaller: Indicates that researcher's hypothesis would be supported
Marketing dependent variables
Sales Volume
marketing dependent variable
Sales volume can be categorized as a ________________________.
Ratio
Sales, price, number of customers, market share, weight, height
Ordinal Scales
Scale with mutually exclusive and collectively exhaustive categories, as well as the property of order, but not distance or unique origin
Nominal scales
Scale with mutually exclusive and collectively exhaustive categories, but without the properties of order, distance, or unique origin
Interval Scales
Scales with the properties of order and equal distance between points and with mutually exclusive and exhaustive categories
How to start questions
Start topics with easier questions, followed by more complex questions Order topics from easier to answer to more difficult to answer
Types of target questions
Structured Unstructured
Methodology
Summarize sample design (including sample profile, margin of error and significance level) Describe research design How are data collected and rationale for using multiple methods for collecting data Data analysis to understand the findings Identify limitations
Conclusion
Thank participants, conclude that participation valuable, purpose of their data help determine.
-1 and +1
The Pearson product-moment correlation coefficient ranges between
Interval
The Pearson product-moment correlation requires what type of scale for the data?
Parallel forms
The U.S. News and World Report is assessing the reliability of its measure of academic reputation used in its annual ranking of colleges and universities. To do so, it administers the same questions on academic reputation to its sample of academic administrators in March and again in June. If the U.S. News and World Report seeks to assess equivalence of alternate forms of measuring academic reputation, it should utilize a ________ type of test.
Test-retest
The U.S. News and World Report is assessing the reliability of its measure of academic reputation used in its annual ranking of colleges and universities. To do so, it administers the same questions on academic reputation to its sample of academic administrators in March and again in June. Which type of test is used in the U.S. News and World Report example to assess reliability?
Pearson's (product moment) correlation coefficient
The ________ is a statistic summarizing the strength of association between two continuous variables.
Scaling
The assignment of numbers or symbols to a property of objects according to value or magnitude is called ________.
Linearity
The assumption that data can be described by a straight line passing through the data array is called ________.
Content validity
The extent to which it provides adequate coverage of the investigative question guiding the study
Constant-Sum Scale
The participant allocates points to more than one attribute or property indicant, such that they total to 100 or 10
Research Objectives
To measure characteristics of the participant in the study To use participants as judges of the indicants of objects or properties presented to them
Pie graph
Uses sections of a circle to represent 100 percent of a frequency distribution of the subject being graphed
Hypothesis testing Procedure
Value is consistent with the hypothesis the hypothesis is supported Value is inconsistent with the hypothesis the hypothesis is not supported
management dependent variables
Variables that include conflict resolution, employee satisfaction, and employee turnover can be categorized as ____________________.
The direction of the relationship
What does the sign (+ or -) of the correlation coefficient signify?
Administrative questions
What type of question identifies the participant, interviewer, interview location, and conditions?
Unidimensional
When a measurement question seeks to measure only one attribute of the participant or object, it is said to be ________.
multiple regression analysis
When a researcher is attempting to predict sales volume by using building permits, amount of advertising, and the income levels of residents, the researcher is using _____.
Equivalence
When an instrument secures consistent results with repeated measures by the same investigator or different samples
increases; increases
When the correlation between two variables is +0.92, this means that as one variable ______ , the other variable ______.
independent; dependent
When the on-time performance of airlines is used to predict the number of customer complaints in a regression equation, on-time performance is the ______ variable and the number of customer complaints is the ______ variable.
Primacy effect
Which form of order bias occurs when a participant tends to choose the first alternative presented as his or her response to a closed-ended question?
Mean
Which measure of central tendency is appropriate for interval scales?
Encourage each participant to provide accurate responses
Which of the following goals should a good measurement instrument accomplish?
multivariate statistical analysis
Which type of analysis involves three or more variables?
Filter question
Which type of classification question determines whether a participant is asked or is excluded from answering one or more questions within a target question topic?
Error of strictness
Which type of rater error occurs when a participant consistently chooses the extreme response alternative at one end of the scale?
Coding scheme
________ involves assigning numbers or other symbols to answers so that the responses can be grouped into a limited number of categories.
Validity
________ is the extent to which a measurement tool actually measures what we wish to measure.
Histogram
a graphical bar chart that groups continuous data values into equal intervals with one bar for each interval
Bar graph
a graphical presentation technique that represents frequency data as horizontal or vertical bars vertical bars are most often used for time series and quantitative classifications
Intra-rater reliability
a measure of the consistency of the application of a coding scheme between raters of context across texts when multiple raters are used
Inter-rater reliability
a measure of the consistency of the application of a coding scheme by a single rater of content, across texts
Filter Questions
a question that determines whether a participant is asked one or more questions within a target question topic or excluded from answering
Rapport
a relationship characterized by agreement, mutual understanding or empathy that makes communication possible or easy
Mapping rules
a scheme for assigning numbers to aspects of an empirical event
Measurement instrument
a sequenced list of questions, crafted using carious scale options, complete with an introduction, section instructions and a conclusion
Bivariate Correlation Analysis
a statistical technique to assess the relationship of two continuous variables measured on an interval or ratio scale shows how much x will change when there is a change in y
Content analysis
a systematic, objective approach used to code message characteristics so researchers can treat diverse textual or verbal content quantitatively as they look for patterns and draw inferences
Boxplots
a visual image of the variables distribution location, spread, shape, tail length, and outliers
Target questions
address the investigative questions of a specific study. Represent the largest number and most important
Stem and Leaf Displays
an exploratory data analysis display for variables with continuous data that uses actual values to group data rather than equal intervals
Frequency table
arrays category codes from lowest value to highest value, with columns for count, percent, valid percent, and cumulative percent
Categorization
asks participants to put themselves or property indicants in groups or categories
Double-barreled question
asks two or more questions but provides the response option to answer only one
Primacy effect
choosing the first option given
Recency effect
choosing the last or most recent option
Ranking scales
constrain the study participant to making comparisons and determining order among two or more properties or objects
Coding scheme
contains each variable in the study and specifies the application of mapping rules to the response codes of each variable
Relational hypothesis
describes a relationship between two or more concepts/constructs
Screen Questions
determine whether a participant has the requisite level of knowledge to participate
Error of leniency
easy raters
Non question elements
enhance rapport
Introduction
first thing they see, eye appealing, time requirements included, encourage participants to answer truthfully
Unstructured Questions
free-response or open-ended questions which give participants the option of choosing their own words
Pareto Diagrams
graphical presentation that represents frequency data as a bar chart, ordered from most to least, overlaid with a line graph denoting the cumulative percentage at each variable level
Error of strictness
hard raters
Balanced rating scale
has an equal number of categories above and below the midpoint
Unbalanced rating scale
has an unequal number of favorable and unfavorable response choices
Dummy tables
helps researchers visualize the data that will be analyzed
Mean
interval or ratio variables
Simple regression
involves finding the best fit line, given the set of observations plotted in a two dimensional space
Unsupported assumption
is one that assumes (rightly or wrongly) an implied affirmative answer to an unasked question in order to answer the current question
Variate
mathematical way in which a set of variables can be represented with one equation
Classification questions
may be free-response, dichotomous choice, multiple choice, or rating questions
frequencies
nominal or ordinal variables
Question coverage
number of questions needs to adequately measure the variable as operationally defined Single question vs series of questions
Unidimensional scale
on seeks to measure only one attribute of the participant or object
error of central tendency
participants tendency to choose the middle option
p-value
probability of observing a sample value as extreme as or more extreme than the value actually observed, given that the null hypothesis is true
Unforced-choice rating scale
provides participants with an opportunity to express no opinion when they are unable to make a choice among the alternatives offered
Ordinal
rank or order- class rank, rating surveys
Multidimensional scale
recognizes that an object might be better described with several dimensions than on a unidimensional continuum
Measurement scale
refers to the type of measurement question a researcher choses to use to collect the data
Forced-choice rating scale
requires that participants select one of the offered alternatives
Sorting
requires that participants sort objects or properties on some criterion
Response alternatives
should cover 90% of participants expected answers
r=-1 or +1
strong correlation
Leading Question
suggests to the participant the desired answer; its biased phrasing makes the question worthless
Halo effect
systematic bias that the rater introduces by carrying over a generalized impression of the subject from one rating to another
Cross-tabulation
technique for comparing data from two or more variables that results in a table
data collection
the collective actions that field a measurement instrument
Measurement instrument
the compilation of measurement questions and non-question elements
Data entry
the process of converting information gathered by secondary or primary methods to a medium for viewing and manipulation
Pearson Correlation Coefficient
the r symbolizes the estimate of strength of linear association and its direction between interval and ratio variables
Rating scale
used when participants score an object, property, or its comparison without making a direct comparision to another object
Regression Analysis
uses simple and multiple predictions to predict y from x values
Measurement questions
what the researcher will ask the participant or record about cases during a research study
Internal consistency
Characteristic of an instrument in which the items are homogenous
Properties
Characteristics of objects that are measured
Structured Questions
Closed questions where the specific response alternatives are provided
Hypothesis testing Procedure
1. Hypothesis is derived from the research objectives 2. A sample is obtained and variable is measured 3. The measured value obtained in the sample is compared to the value either stated explicitly or implied in the hypothesis
· Audience-Centric Planning
1. Provide findings, insights and recommendations 2. Raise concerns about unknown problems 3. Showcase emerging opportunities 4. Convince audience to take action related to research objective
Coding Verbal Responses/ Content Analysis
1. Select the Content you want to analyze 2. Define units and categories of analysis 3. Develop set of rules for coding 4. Code text according to rules 5. Analyze results and draw conclusions
Mapping rule
A ________ is a scheme for assigning numbers or symbols to represent aspects of the event being measured.
Concept
A bundle of meanings or characteristics associated with certain concrete, unambiguous events, objects, conditions, or situations
Reliability
A characteristic of measurement concerned with accuracy, precision and consistency
Validity
A characteristic of measurement concerned with the extent that a test measures what the researcher actually wishes to measure
Variable
A characteristic, trait, or attribute that is measured
Operational Definition
A definition for a variable stated in terms of specific testing criteria or operations, specifying what must be counted, measured, or gathered through our senses
Regression analysis
A dependent variable y is lined to an independent variable x Attempts to predict the values of y a continuous interval-scaled dependent variable from specific values of the independent variable Simple regression involves finding the best fit line, given the set of observations plotted in a two dimensional space
Stapel Scale
A numerical scale with up two categories (half positive, half negative) in which the central position is an attribute. The higher the positive number
Editing
A process for verifying that variables have used the designated coding scheme and that all data collected is entered correctly
Numerical scale
A scale in which equal intervals separate the numeric scale points, while verbal anchors serve as labels fro extreme points
Graphic Rating Scale
A scale in which the participant places his or her response along a line or continuum
Multiple-choice, Multiple Response scale
A scale that offers the participant multiple options and solicits one or more answers
Multiple-choice, single response scale
A scale that poses more than two category responses but seeks a single answer, or one that seeks a single rating from a gradation of preferences, interest or agreement
Visual Graphic Rating Scale
A scale where participants mark their response at any point along a line between two response choices that are represented by images or symbols, rather than words
Ratio Scales
A scale with the properties of categorization, order, equal intervals, and unique origin
Simple category scale
A scale with two mutually exclusive response choices
Multiple Rating List Scale
A single interval or ordinal numerical scale where raters respond to a series of objects
Cross-tabulation
A statistical technique that describes two or more variables simultaneously and results in tables that reflect the joint distribution of two or more variables that have a limited number of categories or distinct values is a(n) ________.
Likert Scale Summated Rating
A variation of the summated rating scale, this scale asks a rater to agree or disagree with statements that express either favorable or unfavorable attitudes toward the object. The strength of attitude is reflected in the assigned score, and individual scores may be totaled for an overall attitude measure
Operational defintion
A(n) ________ defines a variable in terms of specific measurement criteria.
Internal validity
Ability of a research instrument to measure what is purported to measure
3 Question Types
Administrative Target Classification
Information
Aggregated Organized Applying rules of measurement Developing data visualization
Evaluating reliability estimates
All of the following activities except ________ are included in the data preparation process.
Assuring confidentiality Building participant interest in the topic Offering an incentive
All of the following can improve participant motivation to participate except ________.
Construct
An abstract idea specifically invested for a given research and/or theory-building purpose
Multiple Regression Analysis
Analyzing effects of two or more independent variables on a single, interval-scaled dependent variable Multiple independent variables include: price, seasonality, interest rates, advertising intensity, consumer income, other economic factors
Insights
Analyzing information Deduction or induction Actionable insights
Number of Scale points
As the ____ increases, the reliability of the measure increases May produce more valid results
Audience Analysis
Audience composition Research question knowledge Research predispositions Recommendation effects Audience effect Supplemental purposes
Balanced vs. Unbalanced
Balanced - same relative weight on each side of the frame Unbalanced - more relative weight on one side
Multivariate analysis
Business research studies involve many variables that must be organized for meaning Business problems involving this include: employee motivation research, customer psychographic profiles, market segment identification research
Selecting a Measurement Scale
Can influence reliability, validity and practivality
Stability
Characteristic of a measurement scale if it provides consistent results with repeated measures of the same person with the same instrument
Exploratory
Data analysis is ________ when patterns in the collected data guide the data analysis or suggest revisions to the preliminary data analysis plan.
Outliers
Data points that exceed plus 1.5 of the interquartile range are called ________.
Dispersion
Describes how scores cluster or scatter in a distribution
Transition
Descriptive statements that are designed to ask participant to a change in topic or a shift in question group
Research Report Plan
Desired audience effect Report structure Content and style
Pearson's product-moment correlation
Determines the degree to which a relationship is linear Measure of strength and direction of association between two variables No significant outliers Normally distributed Linearity Indicates how far away data points are from line of best fit Interval or ratio variables
3 Phases Developing the Measurement Instrument
Develop and refine the measurement questions Develop and refine the non-question elements Develop and refine the measurement instrument
Desired Audience effect
Earning the audience's embrace of findings, insights, and recommendations Connect to the audience Prep for the possibility of unreceptive or unsupportive audience reaction
Number of dimensions
Either unidimensional or multidimensional
Measurement instrument should accomplish
Encourage individual to participate Help establish and maintain rapport with the participant Encourage participant to provide accurate responses Encourage participant to provide adequate information Discourage participant from refusing to answer specific questions Discourage participant from early discontinuation of participation Leave participant with a positive attitude about survey participation
Rater errors
Error of central tendency Error of leniency Error of strictness Halo effect
Question wording
Every word you use in you question has the potential to bias your participant's response
Differences of Standard hypotheses
Examine how some variable differs from some preconceived standard
Differences of groups hypotheses
Examine how some variables varies from one group to another Common in casual designs and group differences in survey research
Data centric
Focus on data and sharing as much data was discovered Factual and statistical
Audience centric
Focus on gaining audience's embrace of data insights and recommendations Persuasive and tells a story Statistics are a tool, not a focus
Forced or Unforced Choices
Forced: requires that participants select one of the offered alternatives Unforced: provides participants with an opportunity to express no opinion when they are unable to make a choice among the alternatives offered
Response Types
Four types: rating, ranking, categorization and sorting
Types of Classification questions
Free-response Dichotomous choice Multiple choice Rating questions
research objectives
Hypotheses are derived from _________________.
Cluster Analysis
Identifies objects or individuals that are similar to one another in some respect Classifies individuals or objects into a small number of mutually exclusive and exhaustive groups Objects or individuals are assigned to groups Similarity exists within groups and not between groups: high internal (within cluster) homogeneity and high external (between cluster) heterogeneity
there is a strong negative relationship between the variables
If the correlation between two variables is - 0.85, this means that ____.
0
If there is no relationship between two variables, the correlation coefficient between them would be
homogeneity; heterogeneity
In cluster analysis, the researcher wants clusters to have high ______ within-clusters and high between-cluster ______.