By successfully completing this course, students will:
- Recognize common distinctions between quantitative and qualitative data and the limitations of such formulations.
- Recognize what is a variable.
- Be introduced to the concept of an experimental controls.
- Be introduced to the logic of random sampling.
- Be acquainted with the concepts of: proportion and its numerical and graphical expression; rate, including rates of change; probability and the nature of randomness.
- Be competent with summary descriptive statistics such as: a mean, a median, a mode, standard deviation, kurtosis and skewness.
- Be able to conduct graphical summaries of data and data visualization.
- Be able to use key measures of association and distinguish the appropriate use of correlation for nominal and interval data.
- Understand the distinction between causation and association.
- Be introduced to linear regression.
- Analyze tabular data of the kind commonly found in reports and understanding how data may be standardized for purposes of comparison.
- Identify sources of measurement error in data and why the misuse of statistics gives it a bad name - obviously to the uninitiated
- Identify common misuses of and mistakes in the presentation of statistics including common fallacies encountered in poor statistical reasoning such as the ecological fallacy.
- Be acquainted to basics in normative and descriptive decision theory, including expected utility and common decision biases
- Gain basic knowledge about formalized decision-making tools
Detailed Course Content
There are a number of aspects of Data Analysis. These are broadly classified as qualitative and quantitative. In this course we focus on understanding of methodological challenges in the analysis and interpretation of data. In the statistical training sessions we will start from elementary concepts of distribution and dispersion to reach an introduction of correlation and linear regression. Decision Analysis is a methodology for ‘objective’ decision making that requires an understanding of probability and game theory. We will debate its merits in seminars as we build capacity in data analysis.
We will be using a statistical package called SPSS. This is the dominant software in business analysis and the social sciences. This is available on the computer labs and there are two sessions scheduled to allow you to familiarize with it. You are welcome to use any other package for doing the statistical analysis, including freeware like R or by employing the statistics module in Excel.
A number of key challenges to data analysts are also explored here. These include method and data triangulation and the debate on the quantitative-qualitative dichotomy.
We will employ sections from a well-respected textbook supplemented by a number of readings of journal articles and chapters from other methodology texts. There will be supplementary reading for those interested in gaining depth. There is a good statistics and methodology section in the library which you are encouraged to use.
In the decision-making unit, basic knowledge is provided about the formal framework in which decision theory is embedded and about several decision-optimizing strategies such as expected utility-maximization, but also about less formal decision-making tools such as multicriteria analysis and performance matrices. A separate chapter is devoted to common decision heuristics and psychological decision biases.