Lecture Notes: Descriptive Statistics and Data Analysis Using Excel

Introduction

• Focus on practicing basic descriptive statistics.
• Use of calculators and software to compute descriptive statistics.
• Discussion on testing causal hypotheses (to be covered later).

Software Tools Available

• SPSS (Statistical Package for the Social Sciences)
• Stata
• SAS
• Excel: Chosen for this semester due to its wide availability.
  • Access through the university's Virtual Computing Lab (VCL) at vcl.ncsu.edu.
  • Contact instructor for access issues.

Understanding Data Sets

• Data sets consist of variables organized in rows and columns.
• Each column represents a variable, and each row represents an observation (e.g., an individual or a country).
• Example Data Set:
  • Variables: Sex, Age, Support for Gay Marriage.
  • Interpretation through codebooks which explain what each number represents.

Practical Example

• Analysis of a public opinion survey with variables such as age, gender, and support for gay marriage.
• Example: Respondent 2 is a female, 23 years old, and supports gay marriage.

Assignment Overview

• Data set focusing on corruption levels by country.
• Scores range from 0 (high corruption) to 100 (no corruption).

Calculating Descriptive Statistics in Excel

• Average:
  • Formula: =AVERAGE(T2:T192) to calculate the mean corruption level.
  • Example outcome: 40.4
• Median:
  • Formula: =MEDIAN(T2:T192)
  • Example outcome: 34 (positive skewness indicated by mean > median).
• Mode:
  • Formula: =MODE.SNGL(T2:T192)
  • Example outcome: 19
• Standard Deviation:
  • Formula: =STDEV.S(T2:T192)
  • Example outcome: 20.89

Interpreting Results

• Summary statistics provide insights into data distribution and variability.
• Positive skewness when the mean is higher than the median.

Handling Missing Values

• In Excel, missing values are represented as "null".
• Different programs like SPSS may display differently (e.g., empty cells).

Next Lecture

• Focus on testing causal hypotheses.
• Example topic: Comparison of corruption in democracies vs. non-democracies.
• Introduction to procedures for hypothesis testing.