Full Summary
Quantitative
Document made to be studied at 150-180% zoom (do this for best experience)
,Lecture 1: Introduction to QM
The importance of Data and QM
Many decisions are based on intuition or rules of thumb (heuristics), not on a rational model.
➢ Heuristics are helpful in strategic management, providing simple ways of dealing with things
o Sometimes it can lead to severe and systematic errors
o Moving from this conceptual entrepreneurship to data-based decision making (with
an experience component).
▪ To make good decisions, you need to develop the right balance between
intuitive and data-based decision making.
You need to know the right data
techniques to use the data, involving
developing and using computer-decision
models.
Trends
- Managers in start-ups have a more data-friendly mindset (than traditional managers)
- Volume of marketing data is exploding
- Many entrepreneurs aware of the importance of having high-powered PC’s
- Firms are reengineering their IT infrastructure for the information age
External and Internal analyses
There are layers and areas of analyses. Macro-environment
Industry: defining a relevant industry
➢ It is the first step for analyzing competitive forces
➢ Example: The street music entrepreneur (factors impacting a street musicians’ success)
Transformation of managerial problem into a research project
,Lecture 2: Essentials of Data Collection and Measurements
Types of data
Secondary data = data already exists within the
company or is collected by third parties for purposes
other than solving the problem at hand (get some initial
insights)
➢ Government publications, Books, Newspapers,
Annual reports, Social networks
It is often worthwhile to check secondary data sources
as a preliminary stage before primary
➢ Sources for new ideas, support the problem, provide a benchmark for checking the validity &
precision of primary data collection, source of methods/techniques for collection & analysis
➢ But, it is incomplete (collected for different purposes), units of measure & level of detail do
not correspond to the requirements, no control over process, data is old
Primary data = data does not exist yet and must be collected by the researcher or third parties
➢ It costs more, takes time and skills, but better targeted
Example of lab experiment: Blind text Coca-Cola (which one they would like better, would go for
Pepsi tasting better when being blind, and Coca-Cola tasting better when seeing)
, Measurements and scales Types of scales
Measurements and behavioral responses
➢ Context matters (pay more for a beer in a fancy resort hotel than in grocery store)
➢ Contrast matters (depends on what other questions are, people compare questions, spray
paint and murder example)
➢ Scale provides cues (how many hours a day
do you spend watching TV)
➢ Language (people use scales different)
➢ Selected error sources of questioning
Lecture 3: Data Editing and Matching & Causal Analyses
Selected examples of data sets
➢ Scanner data = data coming from retail UPC scanners (barcode)
o Potential limitations
▪ Sampling frame (small shops may not be considered)
▪ Cannot make causal statements (right away)
▪ Don’t know behaviors and psychographics
Quantitative
Document made to be studied at 150-180% zoom (do this for best experience)
,Lecture 1: Introduction to QM
The importance of Data and QM
Many decisions are based on intuition or rules of thumb (heuristics), not on a rational model.
➢ Heuristics are helpful in strategic management, providing simple ways of dealing with things
o Sometimes it can lead to severe and systematic errors
o Moving from this conceptual entrepreneurship to data-based decision making (with
an experience component).
▪ To make good decisions, you need to develop the right balance between
intuitive and data-based decision making.
You need to know the right data
techniques to use the data, involving
developing and using computer-decision
models.
Trends
- Managers in start-ups have a more data-friendly mindset (than traditional managers)
- Volume of marketing data is exploding
- Many entrepreneurs aware of the importance of having high-powered PC’s
- Firms are reengineering their IT infrastructure for the information age
External and Internal analyses
There are layers and areas of analyses. Macro-environment
Industry: defining a relevant industry
➢ It is the first step for analyzing competitive forces
➢ Example: The street music entrepreneur (factors impacting a street musicians’ success)
Transformation of managerial problem into a research project
,Lecture 2: Essentials of Data Collection and Measurements
Types of data
Secondary data = data already exists within the
company or is collected by third parties for purposes
other than solving the problem at hand (get some initial
insights)
➢ Government publications, Books, Newspapers,
Annual reports, Social networks
It is often worthwhile to check secondary data sources
as a preliminary stage before primary
➢ Sources for new ideas, support the problem, provide a benchmark for checking the validity &
precision of primary data collection, source of methods/techniques for collection & analysis
➢ But, it is incomplete (collected for different purposes), units of measure & level of detail do
not correspond to the requirements, no control over process, data is old
Primary data = data does not exist yet and must be collected by the researcher or third parties
➢ It costs more, takes time and skills, but better targeted
Example of lab experiment: Blind text Coca-Cola (which one they would like better, would go for
Pepsi tasting better when being blind, and Coca-Cola tasting better when seeing)
, Measurements and scales Types of scales
Measurements and behavioral responses
➢ Context matters (pay more for a beer in a fancy resort hotel than in grocery store)
➢ Contrast matters (depends on what other questions are, people compare questions, spray
paint and murder example)
➢ Scale provides cues (how many hours a day
do you spend watching TV)
➢ Language (people use scales different)
➢ Selected error sources of questioning
Lecture 3: Data Editing and Matching & Causal Analyses
Selected examples of data sets
➢ Scanner data = data coming from retail UPC scanners (barcode)
o Potential limitations
▪ Sampling frame (small shops may not be considered)
▪ Cannot make causal statements (right away)
▪ Don’t know behaviors and psychographics
