4202 makeup exam

interval

- uses distance between objects, zero point is arbitrary ex) temperature, brand attitude -analysis: frequency, mode, median, mean, standard deviation

nominal

- uses identity/category (identical or distinct) ex) marital status, gender, occupation -analysis: frequency, mode is highest frequency

What is your phone number?

nominal

what is your mobile provider?1.Verizon 2. AT&T 3. T-Mobile 4. Other

nominal

exploratory research

plan to gather ideas and insights about the problem

non-probability sampling

subjective procedure in which some members of the population have a zero or unknown probability of being selected - convenience sample -judgment sample - quota samples - there is selection bias here - you do not have a complete sampling frame

rank order scaling (comparative)

order the following attributes in terms of relative importance of each attribute to you 1= highest importance, 4= lowest .......

Please rank the following activities in terms of your preference by assigning numbers: 1 = most preferred, to 5 least preferred. - Reading magazines ... - Shopping ... - Watching television ... - Eating out ... - Dating

ordinal

Rank the following mobile providers according to your degree of liking for each (1=most preferred provider, 3=least preferred provider): Verizon AT&T T-Mobile

ordinal

On an average weekday, how much time do you spend doing your homework and class assignment? - Less than 30 minutes .... - 120 to 240 minutes ... - 31 to 60 minutes ... - more than 240 minutes ... - 60 to 120 minutes ...

ordinal (ranking is meaningful but not the mean or ratio)

casual research

plan to determine cause and effect relationships

descriptive research

plan to determine frequencies and relationships between events

what are the different sampling methods?

probability: -Simple random samples • Systematic samples • Stratified samples • Cluster samples non probability: - Convenience sample • Judgment samples • Quota samples

what is a construct?

a characteristic/ attribute that we would like to measure - concrete example: age - abstract example: attitude toward starbucks

how to order questions to avoid bias and facilitate respondents to answer?

are you happy with your dating life? are you happy with your life? if your dating life isn't going good it can influence to your answer to how you feel about your overall life - people never recall all information - use information that comes first

what is a measurement?

assigning numbers to constructs according to a pre-specified rule -concrete example: my age = 22 - abstract example: my attitude towards Starbucks is 3

parameter (population)

average of population

statistics (sample)

average of sample

semantic differential scale (itemized)

bi-polar scale format that captures a persons feelings about a given object -uses good/bad, unhelpful/helpful -you can report this by a snake diagram

Please divide 100 points between the following attributes in terms of the relative importance of each attribute to you for a bank. Courteous service ...... Convenient location ...... Convenient hours ...... Low-interest-rate loans ...... Total: 100

constant sum

While giving guidelines to her field interviewing staff, a researcher states, "I want you to interview every tenth customer entering a Kroger store at ten key sites." what sampling method is this?

convenience sample, non probability

formulate problem in earth day columbus

decision problem example: should green Columbus continue the earth day Columbus event? research problem examples: what is the level of awareness for promotion campaigns? what is level of participation among different people in Columbus

scaling

generation of a continuum upon which measured objects are located -concrete example: 0-120 years -abstract example: 1= unfavorable attitude 2= neutral attitude 3= favorable attitude

what are the goals of exploratory research?

- understand managers decision problem better - formulate research problem better - develop hypotheses to be tested in descriptive/casual research

ordinal

- uses ranking order (higher or lower) ex) brand preferences, letter grades - analysis: frequency, mode, median is middle point

comparative scale

-allows direct comparison and relative thinking - can't have too many items

what is validity?

-correctness -perfect validity: observed score= true score random error= 0 systematic error= 0 - difficult to access equation: observed score= true score

systematic sampling (prob)

-easier to implement if data is not in a computer ex) She draws one random number n between 1 and 100. Using the list of 60,000 addresses she selects the following students: n, n + 100, n + 2*100, n +3*100, ... , n + k*100, ... , n + 599*100.

non comparative scales

-easy to formulate -easy for respondents to answer - interval scale

how can you achieve the goal of marketing research?

-internal decisions: product, price, place, promotion -external factors: competitors, suppliers, social trends

focus group limitations

-observer interference - be influenced by other people - subjective interpretation of discussion - not representative

ratio

-uses absolute distances (meaningful zero) ex) units sold, weigh, length, price, spending -analysis: all (including ratios of differences; twice as much)

what are two types of constructs in marketing research?

1) concrete (observable) characteristics: age, gender, education, brand last purchased 2) abstract (unobserved/latent) characteristics: involvement, loyalty, attitude, satisfaction

why do we need marketing research?

1) developing and planning 2) implementing and problem solving 3) monitoring the firms marketing plans

what are three research designs?

1) exploratory research 2) descriptive research 3) casual research

how old are you?

ratio

What is a funnel approach? Why do we use this approach?

- start with easy general questions - put important and specific questions in the middle -at the end put sensitive and demographic questions

focus groups

- 6-12 people - homogeneous characteristics - no professionals -moderator is well prepared and keeps the conversation going - location is comfortable and stimulate interaction, online are getting more popular, record focus interview on video

what do you do for a research design to be good?

- be patient: don't start collecting data too early - for each research problem you need one research design

key things in ordinal scale

- can be placed in order - usually based on satisfaction not numbers -

what is reliability?

- consistency in measurement -perfect reliability random error= 0 assessment: use multiple items, between 0 and 1, alpha =1 perfcet, alpha .7 measure considered reliable equation: observed score = true score + systematic error

convenience (non prob)

- convenient special case: voluntary response sample, ex) movie ratings ex) She proposes to go to one of the restaurants at campus and interview 600 students that just had lunch.

key things in nominal scale

- do not have numeric value - have no order - can be called labels - doesn't overlap, mutually exclusive

cluster (prob)

- easy to implement, cost efficient - but it requires subpopulations to be relatively similar ex)Using the 4 groups from method c. She randomly selects 2 groups, and from each of these 2 groups, she uses the list of emails and randomly selects 300 students.

what do you do good for exploratory research?

- focus groups are popular - projective techniques in focus groups

what should be in your introduction and how do you make it interesting?

- give background story, find quotes in newspapers or magazines that shows the problem is important and really is a problem -background and context, - introduce decision problem - introduce research problem

simple random sample (prob)

- highly generalized, easy to understand ex) Using a computer program she randomly selects 600 email addresses from the complete list. • Special case: census

stratified (prob)

- includes all relevant subpopulations - requires subpopulations to be relatively different ex)She divides the students in 4 different groups according to disciplines: business, humanities, engineering, and science. From each group, she gets the list of emails and randomly selects 150 students using a computer program.

key things in ratio scale

- like interval however zero is meaningful - tell us exact value between units and have absolute zero - height, weight can be added, subtracted, divided, and multiplied

what are some exploratory research methods?

- literature search (usually start here) - depth interviews - focus groups - observations - projective techniques

key things in interval scale

- ordered numbers with meaningful divisions - we know order and exact differences between values ex) temp 60 to 50 is 10 degrees - space in-between (value between each item) they don't have true zero - you can add and subtract but not multiply or divide

judgment (snowball)

- possibly only alternative - very subjective ex)She proposes to interview only students that live on campus and have at least 5 dinners a week. Fortunately, she knows some students have dinner at campus every day. She suggests to interview those students first, and to ask these students to suggest other candidates that have dinner at least 5 times a week.

What is the problem in the following survey questions? - To what extent do you agree: "Spectrum's internet connection is fast and reliable." - Which brand of soft drink do you drink? - What is your annual income (before tax)? .......

- question 1 is two questions - too broad - too personal and specific

what is a conceptual model? where should you use it?

- representation of a system, make composition of concepts which are used to help people know, understand, or simulate - summarizes exploratory research and tells us which relationships we want to test in further (descriptive or casual research)

quota (non prob)

- sample controlled for characteristics ex)Because 3/4 of all students at OSU are undergraduate students, she suggests to interview 450 undergraduate students and 150 graduate students that just had lunch at campus.

how do you look for good research problems?

1) exploratory research: literature study (which promotion tools do similar events use) focus groups, interviews 2) descriptive research: surveys (what is the level of participation? what is the level of awareness?) 3) casual research: test cause and effect (test two different e mail promotions) - look at it from researcher perspective, be specific, information oriented, and answer more objective - order research problems logically - first part is usually exploratory with secondary data first then primary -second part of research problems are descriptive or causal

research is a process, what are those steps?

1) formulate problem 2) research design 3) data collection 4) analyze data 5) present results

what are the different formats of scales?

1) itemized - likert scale differential scale - semantic differential scale 2) graphic - slider scale 3) comparative - paired comparison scaling - rank order scaling - constant sum scale

who spends the most on marketing research?

1) nielsen 2) kantar 3) quinities IMS

types of scales

1) nominal 2) ordinal 3) interval 4) ratio

what categories of firms conduct marketing research?

1) producers of products and services 2) advertising agencies 3) marketing research companies

what are three requirements necessary for a casual relationship?

1) relationship between x and y 2) time order: first x than y 3) elimination of other possible factors

two types of error:

1) systematic (constant): due to personality traits, stable characteristics, consistent answers across time, can be adjusted potentially 2) random error: temporary, situational factors, inconsistent across time, cancel out across respondents observed score= true score + systematic error + random error

paired comparison scaling (comparative)

1) which of the following attributes of a bank matters to you? courteous service or convenient location

What would be the impression score toward the bank for the above participant? (using 1 to 5 to represent bad to good impression)

Answer: (5+2+4+4)/4

1. Your name? ........... 2. Age? ........ 3. Income? .............. 4. Below are different statements about Spectrum's internet cable service. Please indicate how strongly you agree or disagree with each. 5. What is your overall satisfaction with Spectrum's Internet cable service? dissatisfied to satisfied

Delete q1 • Order should be 5-4-2-3 (from general to specific to demographics) • Age and income consider change answers to categories • Income specify yearly/monthly • q4.2 double barreled, ask two questions separately • Things to consider changing depending on your participants' knowledge: - q4.4: "modem" may need to be explained - q4.3: ask a screening question first that they have experience to pay on website

how to formulate questions appropriately?

If possible: close ended questions instead of open-ended questions (Likert or semantic differential scales) • Limit the length of the questions Rules of thumb: - < 15 words - < one full line in print • Use an efficient lay-out • Be very specific about the product/service when asking about purchase likelihood - Detailed features of the laptop • Be very specific when referring to time - Often used duration: two weeks / one month

Why does a Likert scale, semantic differential scale, an exam or an intelligence test consist of a number of test questions?

Increases reliability - Increases validity

constant sum scale (comparative)

Please divide 100 points between the following attributes in terms of the relative importance of each attribute to you. Courteous service ...... Convenient location ...... Convenient hours ...... Low-interest-rate loans ...... Total: 100

funnel approach

Start with easy general questions - Screening questions • Do you have a Netflix account? - Start with interesting questions (opinions) • Which of the following streaming services have you used in the last 30 days? • In the middle - Important questions that are more difficult (from general to specific) • Overall, I enjoy the service provided by Verizon • Verizon charges fair prices • At the end - Sensitive and demographic questions • Age, sex, education, income, occupation • Why funnel approach? 1. Make respondent feel comfortable and committed 2. Avoid order bias

probability sampling

each member of the population has a known, and non-zero probability of being selected - simple random sample - systematic samples - stratified samples - cluster samples - it is difficult to construct sapling frame

how do you proof causality?

experiment

what type of validity is high in field experiments?

external validity (executed in realistic situations)

what is the most important step in marketing research process

formulate problem (step one): if you make a mistake in this first step everything else if ruined

why is the first step of the research problem so important?

formulate problem: - you determine decision problem - you determine research problems

the type of experiment (laboratory or field) has major implications on what?

internal and external validity

what type of validity is high in laboratory experiments?

internal validity because the researcher has more control

I am satisfied with Verizon's network 1=strongly disagree, 2= disagree, 3=neither agree nor disagree, 4= agree, 5 = strongly agree

interval

I like to solve crossword puzzles (please circle your answer) Disagree Agree 1 2 3 4 5

interval

compute attitude score with multiple items:

measures respondents attitude towards more than one attribute q1+q2+q3+q4+q5/5 be careful because if its opposite of what the question is you will count starting from the right and not left

slider scale (graphic)

not important to extremely important put circle on where it lies

A researcher wants to study the effects of social class on consumption behavior. He establishes three different categories of social class in terms of amount of income. He then assigns each field worker a specific number of interviews with people in each income category. The interviewers are allowed to select who they interview.

quota sample, non probability

How long have you been using Verizon? years

ratio

How much money did you spend last month on entertainment? .....

ratio

laboratory experiment

researcher creates environment (ex: marketing experiment for monetary rewards)

field experiment

researcher uses a realistic situation (example: A/B testing)

likert scale (itemized)

scale that asks respondents to indicate the extent to which they agree or disagree with a series of statements about a given object. strongly disagree to strongly agree

Please check the space that best represents your opinion for Huntington bank. discourteous to courteous inconvenient to convenient

semantic differential scale

what should and shouldn't a moderator do?

should: use probes like tell us more about that, ask specifics, get everyone involved, get them and keep them back to focus, covers the 4 p's (will help stimulate a discussion) should not: ask leading questions

what is marketing research?

the process of gathering and interpreting data for use in developing and planning, implementing and problem solving, and monitoring a firms marketing plans

sampling

the selection of a small number of elements from a larger population defined target group - this draws general conclusions about the target group

what is the goal of marketing research?

to create exchanges that satisfy the needs of both customers and marketers

what is internal validity in an experiment?

to what extent can we be sure that X (loyalty program) really causes y (spending)?

what is external validity?

to what extent can we generalize the results of the experiment to other situations?

projective techniques

usually used if consumers are giving socially acceptable answers and hiding their true feelings use: association or sentence completion