Research Techniques Flashcards
Observational research has an issue, which?
1) They might have a bias on their observations
2) Might introduce bias if customer change behavior due to their presence
3) Sensitive to external influences (like weather etc.)
What can systematic market research be used for?
- Gathering and analyzing relevant data allows for developing a better understanding of markets and customers
- The effect of specific marketing measures can be assessed, informing decision-makers about necessary adjustments
- Information can be based on qualitative (e.g. interviews) and quantitative evidence (large-scale data); primary and secondary research
Mention the 5 common research approaches in market research
- Observational Research - Ethnographic research, observations of consumers in store. Goal is to immerse the researcher into consumers’ live to uncover unarticulated desires, meanings or behavior
- Focus groups - Assembling a small group of individuals with intention to discuss in-depth various topics of interest, use of moderator. Mainly for exploration
- Surveys - Learning about people’s knowledge, beliefs, preferences and to measure magnitude thereof in population
- Behavioral & big data - Customers leave traces of purchasing behavior (e.g. customer databases, in store scanning data), subsequent (big) data analytics to identify patterns
- Experimental Research - Capturing cause-and-effect relationships by eliminating interfering factors. Most reliable insights (but limited generalizability)
What are the advantages of each common research approach?
Expensive, Limited information, Difficult to control unforeseen factors
2. Focus group – Richness in data, fast data collection – Downsides: Limited generalizability, relies on skills of interviewer, difficult to replicate findings
3. Survey – Inexpensive, High reliability and generalizability, replicability – Downsides: Potential survey errors, need expertise to create a proper survey, low response rates
4. Behavioural and big data - Limited subject bias, relates to “real” behaviour, high reliability of behaviour – Downsides: Consumer privacy, difficult to handle data, needs specialized analyst
5. Experimental research – May capture cause-and-effect correlation, control over variables, replicability. Downisdes: Artificial, ethical concerns, costly in time and money
What are the two way of sampling in a survey?
1) Probability sample – Selection of individuals based probabilities of selection (divided into three types)
2) Non-probability sample – Selection individuals based on criteria (divided into three types)
What are the three types of Probability sampling?
1) Simple random sample: Every member of the population has equal chance of being selected.
2) Stratified random sample: Population is divided into mutually exclusive groups (like age) and random samples are drawn from each group.
3) Cluster (area) samle: Population is divided into mutually exclusive groups (like city blocks) and researcher draws a sample of the groups to interview.
What are the three types of Non-Probability sampling?
1) Convenience sample: Selecting the most accessible population members (like students)
2) Judgement sample: Selecting population members who are good prospects for accurate information
3) Quota sample: Finding and interviewing a prescribed number of people in each of several categories
What 6 common possibilities are there in Questionnaire Design?
1) Dichotomous: A question with two possible answers (Yes/No)
2) Multiple choice: Question with three or more answers
3) Likert Scale: Statement with amount of agreement/disagreement (from 1-5, sometime 1-4)
4) Semantic differential: A scale connecting two bipolar words and a selection somewhere on the scale of participant. (Like Small/Large)
5) Importance scale: A scale that rates the importance of some attribute (1-5/Extremely important to not at all important – “3G cell service is…”)
6) Intention-to-buy scale: A scale that describes the respondent’s intention to buy (Definetly buy-Probably buy-Not sure-Probably not buy-Definetly not buy)
What is a conjoint analysis?
A special questionnaire that identifies consumer preference using discrete choices and assigning value to different product attributes.
1) Potential customers are presented different combinations of product characteristics (observable bundles) and asked to indicate their preferred one
2) Attributes are assign value: Which attributes are most influential on the choice of the respondents? Each product attribute will receive some “value”: does screen or sound quality matter more for buyers of new TV’s?
3) Product attributes can now be made explicit without consumers having to state their most important criteria explicitly. Advantage is the similarity to actual buying situation (as full products are evaluated).
Give an example of a conjoint analysis
Phones: Price ($$$ vs. $) - Screen size (Small vs. big) - Camera resolution (2mpx vs. 6mpx)
What levels can you decompose the market of a product into?
1) Potential market: set of consumers with a potential interest in a market offer
2) Available market: set of consumers with interest, income and access to a particular offer.
3) Target market: part of the available market the company decides to pursue (c.f. relevant market segments)
4) Penetrated market: set of consumers buying the company’s product
What is TAM
Total Addressable Market - The overall opportunity that is available to a product or service if 100% market share was achieved
What is SAM
Serviceable Addressable Market - Serviceable addressable market is the part of the total addressable market (TAM) that can actually be reached
What is SOM?
Serviceable Obtainable Market - An estimate of the portion within a specific market segment that a company is able to capture. So the part of the SAM that can be captured.
How can you estimate a total market?
1) The chain ratio method
2) The market-buildup method
Explain the chain ratio method for calculating total market size
- Starting with a base number and adjusting it for several percentages that reflect relevant sub-groups / segments
Overall outcome depends on accuracy of shares at every stage, errors magnify multiplicatively
Example: 250 mio people in the US x 0,2 (20% have diabetes) x 0,02 (2% get treatment) x 0,5 (50% market share for Novo) = Total Market
Explain the market-buildup method for calculating total market size
Relying on formal (sub-)sector definitions to measure demand – measuring demand through databases
- Easy to gather through databases such as ORBIS/OSIRIS or Compustat
- Not all firms report accurately their sales de-composed by market segment (c.f. large diversified firms)
What are three main indicators of products/marketing?
1) Corporate Financial Metrics
2) Customer Metrics
3) Advertising/Promotion Metrics
