DATA SOURCES 1. Data comes from many sources: company databases, government reporting, market research, etc.
Volume of data is growing enormously because of
Electronic Data Capture - doubling every two years Electronic capture of data means no major concerns with quality of data / measurement error The challenges: linking data about one subject / person from different sources. This is why loyalty programs / membership / logins are so valuable.
DATA SOURCES 2. Surveys
Extracting targeted information
Allows us to relate information together Discovery eg characteristics and preferences We learn directly from the people we are most interested in. eg poor farmers in a developing country describe their situation and difficulties
DATA SOURCES - SURVEYS
Challenges with surveys 1. Questions that extract the right information
do not have ambiguous meaning
impartial
wording that does not invite dishonesty
DATA SOURCES SURVEYS,
CHALLENGES 2. Samples that are representative
Surveys invariably only include a sample from
the wider population The sample needs to be representative A non-representative sample is Biased, and gives biased statistical results.
SURVEYS, CHALLENGES, BIAS
Sources of bias? 1. Selection Bias: the sample is chosen in a way that makes it not representative. Solution? Taking a random sample 2. Non-Response Bias: not all who are selected for a sample will respond. If there is a pattern to the non-responses, this creates a bias in the sample.