Consumer Insights 201 Final

Interval and ratio-level data are considered ____________ variables and can accommodate numerous types of descriptive statistics

"continuous/numeric"

Name one way to lessen concerns about disguising research?

1. By letting respondents know the study is "blind" and why 2. Allowing respondents to "opt out" 3. Debriefing: providing respondents with details of the research after data have been collected

Name one of three common nonprobability sampling techniques

1. Convenience Sample-Being included in a sample as a matter of convenience, right place, right time. Examples are mall intercepts, surveys on websites 2. Judgement Sample- Sample elements are handpicked because they are expected to serve the research purpose. Examples are snowball and onsite panels. 3. Quota Sample-Sample is constructed with certain characteristics that reflect the target population. Online panels are examples.

Name two tasks in the data preparation stage to ensure quality standards are met

1. Convert all responses to consistent units, e.g. months to years, days to weeks, cents to dollars 2. Assess degree of nonresponse, delete record if >/= 50% are missing 3. Check for consistency across responses 4. Look for evidence that the respondent wasn't thinking about answers, e.g., straight lining" (all 5's) 5. Verify that branching questions were followed correctly, e.g., "if yes, continue to 3" "Otherwise skip to 5"

What are the two ways observation research can be carried out?

1. Direct - watching actual activity 2. Indirect - observing the outcome of activity

What could you do if greater precision is desired

1. Increase the sample size (if possible) 2. Decrease the degree of confidence, such as from 95% to 90%

What are the three MUST HAVES for effective communication

1. KNOW YOUR AUDIENCE 2. USE A "GOLDILOCKS" LEVEL OF DETAIL - NOT TOO MUCH, NOT TOO LITTLE 3. ALWAYS END WITH A VERY CLEAR CALL TO ACTION

Name and briefly define two of the four types of scales used in Primary Research?

1. Nominal - Identification. Numbers don't mean anything. 2. Ordinal - Order. Numbers are meaningful 3. Interval - Comparison of the size of the differences. Attitude measurement. 4. Ratio - Comparison of absolute magnitude, includes natural zero. Ranking.

Name and briefly describe one method of conducting Primary Research

1. Personal Interviews - can take place anywhere, can collect very detailed information 2. Telephone Interviewing - becoming increasingly difficult, due largely to mobile (no lists, can't due RDD). 3. Mail (Paper) Surveys - historically the leading from of primary research. The shift to paperless is one reason for downward trend. 4. Online Surveys - administered online such as through a website, the predominant method

What are the two approaches to drawing a sample and give one example of each?

1. Probability sample: A sample in which each target population element has a known, nonzero chance of being included in the sample.‒ Examples: Simple Random, Systematic, Stratified, and Cluster sampling (including area) 2. Nonprobability: A sample that relies on personal judgment in the element selection process.‒ Examples: Convenience, Judgment, and Quota sampling

Name three ways online surveys can be further optimized?

1. Short as possible 2. Optimize for mobile phones 3. Simple questions 4. Use of visuals/graphics 5. 3D technology or AR as appropriate 6. Show respondents see the progression 7. PRETEST 8. Highlight incentives 9. Gaming techniques 10. Keep it interactive

What are the two key considerations in data analysis?

1. What type of analysis will be done? -Univariate analysis: One variable is analyzed at a time. Major purpose is to describe. -Bivariate or Multivariate analysis: Two or more variables analyzed together. Purpose is to understand relationships 2. What level of measurement will be used? -Four levels of measurement based on the data -Scales each lend themselves to different analyses

What is meant by a filter question and give an example?

A question used to determine if a respondent has the knowledge or "qualifies" for the study Examples: "Do you do the grocery shopping for your family?""Have you shopped at Trader Joes within the past six months?""Did you vote in the last presidential election?"

How is a Chi-Square Good-of-fit Test used and with what kind of variables?

A statistical test on categorical variables to determine whether some observed pattern of frequencies corresponds to an expected pattern.

What is meant by target population and give two examples?

A target population comprises all individuals or entities that meet the designated criteria to qualify for a research study. Examples: ‒ College students ‒ Primary grocery shoppers ‒ Household with kids 6-12

Sample Control

Ability to project or be representative

Advantages and Disadvantages to High Structure

Advantage-Easy to code, organize and analyze Super reliable and easy to make stats Disadvantage-Bias by making people choose through alternative Can be not as precise

What is the primary goal of Descriptive Statistics?

Aim is to describe properties of data based on measures central tendency and of dispersion. Central Tendency: central point of distribution Variability: how the scores are scattered around the central point

Match the scale with the correct statistic

All summary statistics = Ratio Mean = Interval Median = ordinal Mode = Nominal

What is a one-sample t-test and what kind of variables are involved?

Based on continuous variables, compares the mean of the sample data to a known value, such as a population or historical mean

Low Structure Questions

Best when the goal is hearing from consumers in their own words Appropriate when responses are not clear cut More suited to emotional questions

Why is sampling error considered less troublesome?

Can reduce it by increasing sample size Can account for it by calculating margin of error (assuming probability sample).

What is the difference between a Census and a Sample?

Census-Collecting data from all members of a population. Parameters are characteristics or measures of a population. Sample-A subset of individuals or entities from a larger group. Statistics are characteristics or measures of a sample.

Why is knowing your audience SO important?

Comfort level with technical/analytical content? Involvement in the project? Interest in the project? History with the project? Relationship with the team?

What is the Two-Box Technique?

Converting interval-level ratings into a categorical measure, such as for presentation purposes, by only showing percentage of respondents choosing one of the top two positions on a rating scale.

Administrative Control

Costs of sending out survey, such as paper and stamps

Most primary research is ______________in nature

Descriptive

Level of Disguise in a Survey

Disguise is essential when knowing the purpose or sponsor is likely to bias answers Interruption, spontaneous answers, change your answer Also useful when recreating the natural environment Particularly in research Can help get the truth, but still raises concerns It can be a violation of respondents right to know Debriefing provides respondents answers at the end of the survey '

The three types of primary research are:

Exploratory, descriptive, casual

When it comes to the standardization of questions in primary research, with __________________ both the questions and responses are consistent where as with ________________ only the question is consistent.

Fixed alternative Open-ended

What does survey research do?

Helps find the answer to Why? Plays a complementary role to observational research

Name two considerations with closed-ended questions?

Include a "don't know" if applies to a sizeable portion (>/+ 20%) Responses must be exhaustive, may need to include an "other"option Responses must be mutually-exclusive Response order bias occurs when responses to a question are influenced by the order

What is a meant by Snowball Sampling and when might it be used?

Judgement sample that is used to sample special, hard to find, populations in which an initial set of respondents are located and asked to others with same, special characteristics.

Besides knowing your audience, what is another fundamental rule in research presentations

KNOW YOUR STUFF

High Structure Questions

Limited number and clear cut Obtaining factual information and assessing opinions about issues Used to collect ratings in attitudes and perceptions

Information Control

Managing interviewer bias Level of anonymity

Online Surveys

Many abilities due to technology Countless ways to administer Flexibility Quick and responsive

When it comes to hypothesis testing, we are always looking to reject the ____________ hypothesis in favor of the ____________ hypothesis.

NULL ALTERNATIVE

Name three sources of error besides sampling error?

Noncoverage - failure to include qualified elements of the defined population in the sampling frame Nonresponse - failing to obtain information from some of the sample elements of the population. Refusals and Not at homes Response - individual provides an inaccurate response, consciously or subconsciously Recording - Errors due to data editing, coding, or analysis errors. Administrative - Mistakes made by people or machines.

What are the two forms of hypotheses testing?

Null hypothesis and alternative hypothesis

The degree of nonresponse is a good indicator of ________________

Overall quality The response rate itself is also a good way to check the quality of a survey

What is one of the most important things you can do before launching your questionnaire?

PRETEST PRETEST PRETEST

What is the difference between precision and confidence?

Precision-The degree of ERROR in an estimate of a population parameter. Confidence-The degree to which one can feel confident that an estimate approximates the TRUE VALUE. As the precision gets smaller our confidence gets larger

What is the difference between validity and reliability of measures?

Reliability: Results are consistent. How well the measure obtains consistent scores across time or situations Validity: Results satisfy the objectives. How well a test measures what it is supposed to measure.

What are THREE ways to improve response rates?

Respondent interest in topic Survey length Guarantee of confidentiality or anonymity Interviewer characteristics and training Personalization Response incentives Follow-up surveys

What is a sampling frame and give one example?

Sampling Frame is a LIST of population elements from which a sample will be drawn; could be geographic areas, institutions, individuals, or other units. Examples: • Customer database • Member directories • Lists developed by data compilers

Why is Probability Sampling the preferred approach?

Sampling error can be estimated and inferences can be made about the population.

Data analysis should always begin with a ________________________

Summary of descriptive measures, or univariate statistics

An error in measurement that is also known as constant error because it affects the measurement in a constant way is called__________________________?

Systematic Error

The goal in sample surveys is to decrease ___________ error, not any one source of error.

The goal in sample surveys is to decrease ___________ error, not any one source of error.

What is meant by research reports serving as an archive?

The research report is often the ONLY documented history of the project that lives on

What is the key difference between univariate and multivariate analysis?

Univariate analysis is based on one variable at a time (or describing data), with Multivariate Analysis involves more than two variables at a time (or understanding relationships).

primary research involves collecting data by "asking" or _______________ people

Watching

What is the p-value in statistics?

a p-value that is less than the significance level indicates the results are statistically significant Is used with many statistics in both univariate and multivariate analyses to determine if results are meaningful the probability of obtaining a given result if the null hypothesis were true in the population

Which of the following is not a best practice regarding the wording of a question? a. Avoid simple words, err on the side of simplicity b. Avoid ambiguous words and questions c. Avoid leading questions d. Avoid assumed consequences e. Avoid double-barrel questions

a. Avoid simple words, err on the side of simplicity

Personal Interviews

are very useful and distinct from other methods Face to face Can be conducted anywhere Can handle a variety of info Usually open ended questions, but still can use more structured questions Strong sample control Very flexible Time and cost intensive

Which of the following is not a consideration when developing the content of questions? a. Is the question necessary? b. Is the question too revealing? c. Are several questions needed instead of one? d. Do respondents have the necessary information? e. Will respondents give the information?

b. Is the question too revealing?

Which of these strategies will not help lessen respondent nonresponse? a. Guarantee anonymity b. Place sensitive questions right up front c. Include statement showing the situation is NOT unusual, called counter biasing d. Ask in terms of "other people," e.g., Do you think most students cheat? e. Ask for general rather than the specifics (e.g., income categories vs specific income)

b. Place sensitive questions right up front

Which of the following is not true about observation research? a. Less versatile, behavior in the moment, can't observe past or future intentions b. Reveals the "why" behind behaviors c. Not as efficient, is on their terms, may need to "wait" for behaviors to happen d. Highly accurate and less subjective as observation is the "real deal"

b. Reveals the "why" behind behaviors

Which of the following does not apply to sampling error? a. Can be estimated (assuming probability sample) b. Should be the focus as the single most important error c. Due to chance d. Usually less troublesome than other kinds of error e. Decreased by increasing sample size

b. Should be the focus as the single most important error

Which of the following factors do not influence sample size? a. Precision - the greater the precision desired, the larger the sample needed b. Size of population c. Confidence - the bigger the sample, the more confident that the true value falls within the range d. Variation of the characteristic - the greater the variability in the sample the larger the sample size needed

b. Size of population

Which of the following is not a characteristic of online surveys? a. A method of administration that relies on the web for completing the survey b. Explosion in use over the past decade c. Response rates are often very low d. Good flexibility; visuals and complex material possible e. Usually quick and inexpensive

c. Response rates are often very low

Which of the following is not true about highly structured questions? a. most useful when possible replies are known, are limited in number and are clear cut b. work well for obtaining factual information and assessing opinions about issues c. are best suited to understanding motivations, or the reasons "why" d. can be used to collect ratings on attitudes, intentions, awareness and so on. e. all of the above

c. are best suited to understanding motivations, or the reasons "why"

A Chi Square statistic is used to test relationships between __________________ variables, referred to as Tests of Independence in cross tabulation

categorical

Nominal and Ordinal scales are often used to group respondents or objects into categories and are considered ___________ variables or measures

categorical

Which of the following characteristics are not true of correlation? a. A statistic that indicates the degree of linear association between two continuous variables b. The correlation coefficient can range from −1 to +1, the closer to 1 the stronger the association c. Assesses the degree to which the variables change consistently (or not) across cases d. Indicates causality e. The p-value is used to decide significance

d. Indicates causality

Besides calculating the mean and standard deviation, which of the following are common approaches to Descriptive Statistics a. Frequency analysis b. Median split c. Cumulative % breakdowns d. Two-box technique e. All are common approaches to Descriptive Analysis

e. All are common approaches to Descriptive Analysis

Which of the following are the best strategies for handling nonresponse? a. Identify cases with a significant amount of item nonresponse and eliminate completely during the editing process b. Eliminate the case with the missing item(s) from further analyses c. Substitute values for the missing items d. Contact the respondent again e. All are helpful strategies

e. All are helpful strategies

What are the primary uses of Frequency Analysis? a. Univariate categorical analysis, such as Cross tabulation b. Identify blunders and cases with excessive item nonresponse c. Identify outliers d. Identify the median e. All are primary uses of Frequency Analysis

e. All are primary uses of Frequency Analysis

Which of the following is true about Independent samples t-tests? a. Determines whether two groups differ on some characteristic that is assessed on a continuous measure b. Test the difference in means between two variables, an outcome continuous variable and an independent categorical variable c. P-value is used with the test statistic to help decide the outcome d. All are true about independent samples t-tests

e. All are true about independent samples t-tests

Which of the following is true about precision? a. Reflected by the width of a confidence interval, how precise we are with our estimate b. More precision means a more narrow estimate, or range c. Estimated by the sample error. d. Desired precision of an estimate is also sometimes called the allowable or acceptable error in the estimate e. All are true about precision

e. All are true about precision

Which of the following are true of regression? a. Seeks to identify the "best fit" between the predictors and the outcome b. Produces coefficients for each predictor variable that shows the individual impact c. Also produces a model "coefficient of multiple determination" (R2)" that reveals the combined "fit," or how much variation is explained d. Uses the F statistic along with the p-value to determine model significance e. All are true of regression

e. All are true of regression

Preliminary steps in the data analysis phase involve which of the following steps: a. Editing or cleaning the data b. Coding and converting data into consistent symbols c. Running frequency analysis d. Identifying outliers e. All of the above

e. All of the above

Which of the following applies to itemized-rating scales? a. A scale on which individuals must indicate their ratings of an attribute or object b. Is a form of interval scale c. Examples include Likert Summated ratings scales and Semantic Differential scales d. Measures "unobservable" concepts such as attitudes e. All of the above

e. All of the above

Which of the following applies to random error? a. Error in measurement that is temporary b. Usually a result of personal or measurement situation c. Affects the measurement in irregular ways d. Is difficult to control e. All of the above

e. All of the above

Which of the following are considerations in designing scales? a. Number of items in a scale b. Individual versus composite measures c. Odd or even number d. Including a "don't know" or "not applicable" response category e. All of the above

e. All of the above

Which of the following are factors to consider when designing primary research? a. Degree of structure b. Degree of disguise c. Setting d. Method of administration e. All of the above

e. All of the above

Which of the following is true about cross tabs? a. Looking to see whether one variable has an influence on another b. Can also be used to look at the joint distributions of two variables c. Most used multivariate tool for categorical variables d. Also known as "contingency tables" or "two-way analysis" e. All of the above

e. All of the above

Which of the following applies to telescoping error? a. Remembering events as having occurred more recently than they did b. Gets worse as time frame asked about is shorter c. Respondents tend to bring in purchases from broader time frames d. Ideal time frame is 2-4 weeks e. All of the above relate to telescoping error

e. All of the above relate to telescoping error

Which of the following are true of Personal Interviews? a. Involves direct, face-to-face conversation b. Generally strong sampling control (including higher response rates) c. Great flexibility, but higher levels of interviewer bias d. Time- and cost-intensive e. All of these are true

e. All of these are true

Which of the following is not true about descriptive statistics? a. Statistics that describe important characteristics/ properties of the data b. Relies on measures of central tendency and dispersion c. Should be the core of most analyses d. Most commonly used descriptive statistics are the mean and standard deviation e. Involves correlation which adds greater depth of understanding

e. Involves correlation which adds greater depth of understanding

fixed-alternative questions

questions in which the responses are limited to stated alternatives both questions and responses are highly structured Same questions and responses

Open-ended questions

questions that allow respondents to answer however they want Standardized questions but response is open to tend to ask about feelings

Confidence intervals take ____________ into account but do account for other common types of error (e.g., response error, nonresponse error)

sampling error

Regression Analysis is a statistical technique used to understand the influence of a _________________on a ___________variable.

set of independent or predictor variables, dependent or outcome

Degree of structure in a survey

the extent to which the questions are standardized

If we want to be very certain that we capture the true population parameter a ___________ interval is better, though we become ___________ precise in our estimate.

wider less