1/103
Looks like no tags are added yet.
Quantitative methodologies strive for a large enough sample size so results can be projected upon a larger population. True or false?
True
Quantitative research
Results are measured and can be counted
Common quantitative primary research methods
Surveys, experiments, and A/B testing
Surveys
Most common form of primary market research
Surveys are mainly quantitative because they contain mostly closed-ended questions. True or false?
True
Which of the following is a use of surveys that we did NOT talk about in class?
Competitor Survey
In an anonymous survey, even the researcher can’t identify respondents. True or false?
True
Which of the following is the strongest level of measurement?
Ratio
Surveys cannot include open-ended qualitative questions. True or false?
False
Experiments
Examine cause and effect relationships between variables
Statistical tests determine whether the intervention had a significant impact on opinions or behavior. True or false?
True
A/B Testing
A type of experiment often used in digital advertising
New product development (NPD)
Gauges the viability of a proposed market entry, researchers examine optimal features, pricing, potential brand acceptance, data can come from surveys, focus groups, and ethnographic research (or combo of 3)
Brand read/comparison
Assess the brand and its competitors; “Rate your perception of quality regarding the following companies…”
Optimal features
Find out which product features are most important to consumers; “When deciding on your next tablet, how important are the following features?”
Purchase intent
Measure consumers’ likelihood to purchase product/service; “If each of the following tablets were in your price range, how likely would you be to purchase each?”
Pricing
Determine how much consumers expect to pay for the product/service; “How much would you expect to pay for [Tablet 1]?”
Brand perception
Research examines brand awareness and familiarity, key attributes, and what makes a brand unique.
Unaided awareness
Respondents list brands w/o being given a list; measures top-of-mind recall; “Which streaming services are you aware of? Please name any you are aware of.”
Aided awareness
Respondents see a predefined list of brands; “How familiar are you with [insert brand] on a scale of 1-5 with 1 being not at all familiar and 5 being very familiar?”
Descriptors
Respondents provide words or phrases about a brand; “What words or phrases would you use to describe [insert brand]?”
Brand attribute ratings
Respondents provide quantitative scores on key brand attributes; ““Tell me how much you agree or disagree with these statements on a sale of 1 to 5 where 1 means strongly disagree and 5 means strongly agree...” - [Insert company] is focused on quality, - [Insert company] products are comfortable to wear, - [Insert company] is committed to wellness.”
Testing advertising concepts
AKA message testing, evaluates advertisements before full development and production
First impressions
Capture top-of-mind reactions; “What was the main msg of the ad?”
Favorability indicators
Assess overall appeal, comprehension and relevance; “If you came across this ad, how likely would it be to catch your attention?”
Uniqueness and fit
Determine whether the ad stands out while fitting the company’s overall branding; “How diff is this ad compared to ads for similar products?”
Action indicators
Assess whether the ad encourages audience enticement {such as purchasing); “How much does this ad impact your likelihood to purchase this product?”
Customer satisfaction
Collects feedback on current products/services, customer service attributes, and communication/engagement
Customer satisfaction surveys should NOT be sent ASAP after the point of purchase/service interaction. True or false?
False
Trending
Track and compare data over time
Benchmarking
Compare results with peer organizations
Department, unit and location comparisons
Compare scores across functional areas of the same organization
Fielding
Primary research studies during the data collection phase
The 3 most common survey distribution methods are: online, snail mail and _____.
Telephone
Online/mobile distro method
Cheapest, fastest, moderate to high response rate
Snail mail distro method
Moderate price, slowest, very low response rate
Telephone interview distro method
Moderate to high cost, moderate speed, moderate response rate
Online/mobile surveys
Can be distributed through text/email invitations or posted on websites, can include imgs/vids but may be unsuitable when respondents need clarification.
Snail mail surveys
Include a cover letter and a stamped return envelope, can work for specific geographic groups or sensitive topics but are poor for quick turnaround
Telephone surveys
Require trained interviewers, can be useful for B2B research, moderate-length surveys or surveys including screening, but are not ideal for sensitive studies or random general population studies
Send no more than 2 reminders for a survey. True or false?
True
Survey invitations are generally 4-5 sentences long. True or false?
True
Response rates
Help gauge the success of survey deployment, low results may mean some population members are not adequately represented
New product/service development survey response rate
2-10%
Brand perception survey response rate
2-10%
Customer satisfaction response rate
15-30%
Employee satisfaction response rate
40-70%
“Soften the list”
Means alerting the potential respondents ahead of the invitation
Non-response error bias
Occurs when there is a diff btwn those who did complete the survey and those who did not complete the survey
Non-contacts
potential respondents do not receive the request to participate in a survey (meet the criteria but for some reason don’t get survey)
Refusals
Receive invitation but choose not to participate
Self-selection bias
People w/ strong feelings about a topic are more likely to participate
Response bias
Occurs when respondents give answers that misrepresent the truth
Deliberate falsification
Knowingly give false information
Lazy responding/straight lining
Give little thought (randomly rating items) or repeat the same rating over and over (straight lining)
Extremity bias
Favor extreme responses (always 1 or 5)
Interviewer bias
Answers influenced by the presence of an interviewer
Social desirability bias
Give answers viewed as much socially acceptable than the truth
Researchers may access individual’s identifiable information
Confidentiality
Even researchers cannot identify respondents
Anonymity
Identifiable information is not shared outside the research team
Confidentiality
A generic survey link can help achieve this
Anonymity
Information must be carefully safeguarded
Confidentiality
Open-ended questions
Useful when we don’t know answer possibilities, want rich detail, or want to inform future quantitative surveys
Surveys should be limited to 1-3 open-ended questions. True or false?
True
Open-ended questions can be a great way of doing attention checks. True or false?
True
Categorical measurements
Use values that can be counted as parts of distinct groups
Continuous measurements
Use values that can be expressed numerically and placed in a logical order, there is equal distance between each data point
Categorical scales
Nominal and ordinal
Continuous scales
Interval and ratio
Ratio
absolute zero (zero is meaningful)
Interval
distance is meaningful (equal distances)
Ordinal
attributes can be ordered (rank order)
Nominal
attributes are only named (multiple choice or check-a-box)
Uses categories, no natural order to categories (demographic questions, yes/no, multiple choice, checklists)
Nominal
Have 2 or more categories and can be placed in logical/meaningful order (ex. more to less) but the distances between positions is not necessarily equal
Ordinal
Have a clear order and display the exact distance in value along the scale, distances matter, no true zero (0 degrees doesn’t mean there’s no temp)
Interval
Measures have equal distances and include absolute, meaningful zero (age, distance, weight, $ spent, # of purchases)
Ratio
Two types of interval scales
Semantic differential and Likert scales
Semantic differential
Rate between bipolar adjectives (Bad → Good)
Likert
Rate degree of agreement or intensity (Strongly disagree → Neutral → Strongly agree)
Anchored Likert
Only ends are labeled (Very dissatisfied → Very satisfied)
The four scales of measurement from weakest to strongest
Nominal → Ordinal → Interval → Ratio
Ex: Have you watched an Apple TV+ show in the last year? Yes / No
Nominal
Ex: Which shows have you watched? (check all that apply)
Nominal
Rank the following TV shows from best to worst…
Ordinal
Apple TV+ has better shows than most platforms: Strongly disagree → Strongly agree
Interval
How many dollars per month do you spend on streaming per month? Please enter a dollar amount.
Ratio
You should randomize categorical response options without a meaningful order, do not randomize ordinal response scales. True or false?
True
Exhaustive questions
includes all possible responses (Ex. Daily email checks 0-3 / 3-4 / 4-5 / 5+ (overlapping 3,4,5)
Exclusive questions
response options to not overlap (Ex. Daily email checks 1-2 / 3-4 / 5-6 / 7+ (missing 0))
Reverse coding (switching direction of scale item, like positive adjective on the left) helps to identify straight lining when appropriate. True or false?
True
Ambiguous questions
Too vague, open to intepretation
Ex. “Is the professor good?” / “Is the professor credible?”
Ambiguous
Double-barreled questions
Asks two things in one
Ex. “Rate your perception of ADPR 381 and the professor who teaches it.”
Double-barreled
Leading questions
Implies a preferred response
Ex. “How do you rate our exceptional curriculum here at CoJMC?”
Leading
Loaded questions
Pushes socially desirable answers
Ex. AI use question framed by calling AI use inappropriate
Loaded