C207 OA NOTES Module 1/2

Document Content and Description Below

Analytics: is the discovery, analysis and communication of meaningful patterns in data Big Data: is a massive amount of data that is difficult to process using traditional database and software te ... chniques Data Management: includes cleaning and storage of collected data Structured Data: multiple choice Unstructured Data: Searching med notes to see if there was a mention of depression over last 20 years Why is quantitative analysis and analytics important in decision making? Decisions are less subjective, analytics can be explanatory, managers can build predictive models Responses: Variables: experimental units: Relational Database: A database structured to recognize relations among stored items of information  Balanced Scorecard keywords: FLIC = Financial, Learning/Innovation, Internal Business Processes, and Customer; also helps to align to strategy  Simple Index Number – ratio of two values representing the same variable; Presents a Before and After  Simple Composite index – Composes or puts together data from many different sources with no weighing  Weighted Composite index – Applies a weight to some of the data clause BIAS: Reliable Data: is consistent and repeatable/ measure of the instrument (test) Valid Data: measures what is intended to be measured/ does your score represent your ability? Measurement Bias: should be a large enough, representative/random (random sample=lottery) is a prejudice in data that results when the sample is not representative of the population be tested Non-representative sample/ Not random Information Bias: ex: asking irrelevant questions; ex: not recording information we think is relevant and it ends up being relevant later-should record all data and eliminate later what is irrelevant is a prejudice in data that results when either the respondent or the interviewer have an agenda that affects responses Conscious Bias: untrue data: error in data because the questions pushes a certain response Response Bias: untrue data: error because a person might make you feel like you should respond some way Out of Range: outlier is a piece of data furthest away from the mean Systematic Error: Error that you need to do something about to correct Random error: represents errors in data caused by unpredictable statistical fluctuations. Will fix itself Omission Error - Distorted results: occurs when something (for example data or survey response) is missing Three elements of an experimental study: Responses, variables, experimental units This bias occurs if the sample is not randomized. Measurement When an instrument produces a consistent output, it is: Reliable This type of error will cancel itself out over a large number of measurements. Random The use of a placebo that is known by the treatment allocator or researcher is condered a(n): Blind Study Cleaning and storing collected data is: Data Management Types of Data Discrete @ data: only whole values and has clear boundaries. It is not possible to own 3.4 cars; you either own three cars or four. These are discrete data points. Categorical data: includes nominal and Ordinal  Nominal data: - categorical; gender, hair color (name/ color of eyes- can not calculate) o Categories: categorical; Name or Characteristic; cannot calculate o Examples: gender, hair color, age of each, name, color of eyes o Key words: Either or, o Type of data: discrete data/ categorical data [Show More]

Last updated: 3 years ago

Preview 1 out of 14 pages