Module 3 (Lecture 3) Flashcards
(37 cards)
What is secondary data?
Data that already exists within the company or is collected by third parties for purposes other than solving the problem at hand (e.g. books, government publications, annual reports, social networks).
What are three types of external sources in secondary data?
- Standardised data collection ~ by commercial providers (panels among retailers/customers, sales statistics in wholesale).
- Published data ~ by noncommercial providers (publications government bodies, annual reports, uni publications).
- Internet (blogs, competitors’ websites, review sites).
What are the possible uses for a secondary data?
- Providing information at a sufficient level of detail and quality for solving a problem.
- First stage for solving a problem with primary data (can be a source for new ideas, can support the problem definition and the formulation of hypothesis).
What are the potential 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.
What is primary data and in what ways can it be collected?
Primary data does not yet exist and must be collected by the researcher or third parties.
- Observation (with or without survey component).
- Questioning (qualitative or quantitative).
What are the characteristics of qualitative/quantitive questioning? (Primary data)
Qualitative:
1. Unstructured/semi-structured survey.
2. Active role of the respondents.
Quantitative:
3. Structured survey with primarily closed questions.
4. Passive reactive role of the respondents.
What are the typical study goals of qualitative/quantitive questioning? (Primary data)
Qualitative (exploratory approach):
1. Exploration of topics.
2. Deeper understanding of motivations/drivers of purchasing behavior.
3. Collection of sensitive data.
Quantitative (exploratory - descriptive - causal approaches):
4. Monitoring of customer perceptions.
5. Identification and description of segments.
6. Determination of interrelationships.
When is a qualitative data analysis most efficient and what are the limitations?
Qualitative data analysis = most efficient in the early, exploratory state of addressing a problem.
Limitations:
- No representative character.
- No objective measurements since statements must always be in interpreted by the interviewer.
- Combinations of opinions is difficult.
- Limited options for computer based processing
What are the characteristics of without/with survey observations? (Primary data)
Without survey: documentation of the respondents behavior without direct influence of the researcher.
With survey: documentation of the behavioral reaction of the respondent to stimuli.
What are the typical study goals of without/with survey observations? (Primary data)
Without survey (exploratory - causal approaches):
1. Understanding the behavioral processes.
2. Uncovering of interrelationships.
With survey (causal approach):
3. Testing advertising campaigns.
4. Market/lab tests new products.
What is scale?
A scale is a system or framework used to assign values (numbers or categories) to data.
It can be discrete (nominal, ordinal) or continuous (interval, ratio).
What is measurement?
Measurement is how we give meaning to data by assigning numbers or labels based on clear rules.
It can describe amounts (for example €4000 of income) or categories (gender = male).
What is a nominal scale used for? Which statistical methods can be used?
Categorization of objects, e.g. gender, marital status.
Chi-square test.
What is an ordinal scale used for? Which statistical methods can be used?
Ranking of objects in an order, e.g. preference ranking of brands. Differences cannot be measured.
Ordered regression, conjoint analysis.
What is an interval scale used for? Which statistical methods can be used?
A scale where the differences between values are equal, but there is no true zero, e.g. temperature.
T-test, ANOVA, regression, factor analysis.
What is a ratio scale used for? Which statistical methods can be used?
Assigns numerical values to objects, whereby a true zero point exists. You can also say things as “twice as much”, e.g. weight, height. Differences are measurable.
All methods are possible.
Formative vs. Reflective indicators/measurements
Formative indicators measure the reasons for the change of something that is not directly observable (e.g. number of beers). Using a multi-item scale is done to increase the validity.
Reflective indicators measure the effects of the change of something that is not directly observable (e.g. ability to walk straight). Using a multi-item scale is done to increase reliability.
Why is validity important for formative measurement?
Because you need to include all relevant causes of the construct to measure it accurately.
Missing one weakens the meaning of the construct,
Why is reliability important for reflective measurement?
Because reflective indicators are symptoms, they should change consistently together.
If not, the measurement is not trustworthy.
What is over reporting bias?
When people give more positive answers, then they truly feel.
What is interviewer bias?
The presence or behaviour of the interviewer influences the respondent’s answer.
How can question orders affect responses?
Early answers might seem more important.
What is the halo effect in surveys?
When the answer to one question affects responses to next questions.
What is validity?
Scale actually measures what it is intended to measure.
