lecture 3 Flashcards
Secondary data
Data already exists within the company or is collected by parties for purposes other than solving the problem at hand
E.g. government publications, books, newspaper, annual reports, social networks
Review of secondary data
Internal sources (turnover/sales stats, marketing input, cost information, in-house panels among customer and retailers)
External sources (standardized data collection(sales statistics in wholesale, panels among retailers, consumers, joint surveys by competition), published data, internet)
uses and limitations of secondary data
Despite limitations, it is often worthwile to check secondary data sources as a preliminary stage before primary data collection.
Possible uses: Providing information at a sufficient level of detail and quality for solving problem
Preliminary stage for solving a problem with primary data (source for new ideas, support for the problem definition, providing a benchmark for checking validity of primary data, source of methods and techniques for the collection analysis)
Limitations of secondary data:
Data is incomplete because it was generally collected for a different purpose
Units of measure and level of detail of the data do not correspond to the requirements
No control over the process of data collection
Data is too old
secondary (desk) research
Own financial accounts
Database
Competitors online presence and style
Rating platforms
–> secondary data isnt created specifically for us, is already out there and is easy to obtain
Primary data (field research)
Data does not yet exist and must be collected by the researcher or third parties
Is generated specifically for us and our purpose
Is not already out there in the form we require it
It is targeted specifically at our individual needs
Relatively time consuming complex and costly
Ways of gaining primary data (from videoclip)
design a questionnaire
Focus group interview with customers
specific goals on google analytics to track on website how online booking feature is used.
two kinds of questioning
qualitative and quantitative questioning:
Qualitative questioning: Unstructured/semi structured survey; active role of the responsents.
Typical study goals: exploration of topics, deeper understanding of motivations and drivers of purchasing behavior, collection of sensitive data (e.g. alcohol consumption)
Quantitiative questioning: structured survey with primary closed questions, passive (reactive) role of the respondents
Typical study goals: Monitoring of customer perceptions, identification and description of segments, determination of interrelationships (e.g. Purcahse behaviour driver analysis)
Types of primary data collection: Qualitative
In depth/expert interview:
Personal interview with an individual
Advantage: greater willingness to discuss sensitive topics by excluding others, no motivation for social role playing, no pressure by others, sufficient time for every respondent, no counterproductive influencing of ones opinion by others
success factors: establishment of a relaxed and sympathetic atmosphere, ability of the interview to ask intelligent follow up questions, ability to keep the discussion on track
Focus group discussion:
Discussion in small groups
Advantages: stimulattion of the activity through group dynamics, creative interaction between participants, generation of a large quantity of information in a short time, cost advantages compared with individual interviews
Success factors: recruiting of the participants: mix of contrasting and similar personalities,
Moderation skills: attentive listening, flexibility in the guiding discussion
Max 12 participants (less more productive)
3-4 group sessions are sufficient
Quantitative data analysis is most efficient in… + limitations
in the early, exploratory state of addressing a problem
No representative character
No objective measurement since statements must always be interpreted by the interviewer
Aggregation of opinions is difficult
Limited options for efficient computer based processing
Four main types of quantitative interviews
Personal interview, telephone interview, postal survey, online survey
postal survey nd online survey have less number of questions why?
people dont just walk away in person or hang up if theyve already committed
Flexibility of survey methods:
Personal interview:
very high
Telephone interview:
very high
Postal survey:
low
Online survey:
moderate
Duration of interviews quantitatiev data
Personal interview:
long
Telephone interview:
moderate
Postal survey:
long
Online survey:
short
Nonresponse rate for quantitative data
Personal interview:
low
Telephone interview:
moderate
Postal survey:
high
Online survey:
moderate
influence by interviewer (quantitative research)
Personal interview:
possible
Telephone interview:
possible
Postal survey:
very limited
Online survey:
very limited
Observation without survey
Documentation of the behavior of respondents without direct influence of the researcher
Typical study goals: understanding behavioral processes (e.g. searching behavior in supermarkets)
Uncovering of interrelationships (e.g. reaction to advertising message)
Observation with survey
Documentation of the (subconscious) behavioral reaction ofthe respondent to stimuli
Typical goals of the study:
Testing of advertising campaigns
Market/lab tests of new products
Scale
A discrete or continuous space onto which objects are located according to the measurement rules
Measurement
rules for assigning symbols to objects such that these either 1) numerically represent the amount of a characteristics or 2) define whether the object falls into a certain category
scaling and measuring are required actions in order to quantify empirical phenomena
Types of scales
nominal: categorization of objects (no order)
Ordinal: raking of objects in an order, distance between rank objects can not be determined
interval: Assignment of objects to categories, whereby the increments between consecutive categories are identical (Temp scale, satisfaction scale)
Ratio: Assignment of numerical values to objects, whereby a natural zero point exists (comparison of absolute magnitude are possible) (weight, age, salesprice)
Formative indicators (items)
Meausre the resons for the change of a construct that is not directly observable. Creating a multi-item scale serves primary to increase the validity of the measurement
E.g. drunkenness:
Number of beers/shots
Time passed since the last alcoholic drink
Faat content of the last meal
Reflective indicators (items)
Measure the effects of the change of a construct that is not directly observable. Multi-item measurement serves primarily to increase the reliability of the measurement.
E.g. Drunkeness:
Clarity of articulation, ability to walk straight, reaction speed and ability to interact
problems with satisfaction scale
No neutral point due to even numbers of response categories
Primarily positive satisfaction statements (unbalanced)
No exploratory comparison with other products
single item scales + advantage and disadvantages
Measurement of a construct with only one item
E.g. Willingness to buy scale
Advantage: simple, direct measurement
More scales can be measured per survey
Low statistical requirements/restrictions on subsegment analyses
Disadvantage:
Restricted validity of important facets of a construct are not taken into account
Risk of reduced reliability for measurement of a variable that cannot be directly observed
Multi item scale + advantages/disadvantages
Measurement of a complex construct (concept) with multiple facets
E.g.Satisfaction scale
Advantages:
measurement of a phenomenon that is not directly observable
Increased reliability of the measurement for reflective scales
Disadvantages: increased costs, danger of incorrect conclusions if the scale is erroneously defined as formative/reflective
Higher statistical requirements on the development and use of the scales
Assurements and behavioral responses (context matters)
E.g. You are lying on the beach on a hot day. All you have to drink is “warm” water. For the last hour youve been thinking about how much you would enjoy a nice bottle of your favorite beer.
A companion gets up to go and make a phone call and is offered to bring back a beer from the only nearby place where beer is sold.
He says the beer may be expensive and asks how much you are willing to pay for the beer. He says he will not buy the beer if it costs him more than the price you state.
Fancy hotel average: 8
small grocery store: 4
Contrast effect Assurance and behavioral responses
How bad is it to walk across the street with a red light
How bad is it to spray paint in this lecture room
How bad is it to murder your BSDA professor
How bad is it to spray paint in this lecture room
You give different answers to how bad is it to spray paint the lecture room, because it depends on the question asked before
Solutions to contrast effect
randomize questions
scale provides cues (what does this mean) asurements and behavioral responses
Survey one: 16% spend more than 2.5 hours a day watching tv when its the highest category
Survey 2: 62% say they spend less than 2.5 hours a day watching tv when its the lowest category
Asurements and behavioral responses (language)
Native language vs foreign language
Intrinsic association of the language
Languages influence how we use scales, differences in answers due to culture
Selected error sources of questioning (asurements and behavioral responses)
Over reporting (respondents state a more positive attitude than they actually have)
Interviewer bias (personal interview)
Bias because of questions order (first answer choices may be seen as more important than later ones)
Halo effect (contrast effect) (one question and its answer may influence the answer)
Tendency to mark the middle positioning in rating scales
Non-anonymity (anonymity lowers the inhibition threshold to give false answers)
