Applications of Electronic Programs in Economics

Data Analysis Process

• Identify: Establish the questions that need to be answered, and determine why data is needed.
• Collect: Gather data from various sources (internal, external, surveys, interviews, etc.).
• Clean: Ensure data is accurate and free from errors, duplicate records, and formatting issues.
• Analyse: Use statistical techniques to extract insights and identify patterns, trends, and correlations.
• Interpret: Draw conclusions based on the analysis, and make recommendations for action.

Types of Data Analytics

• Descriptive Analytics: Examines historical data to identify patterns, trends, and correlations.
  • Describes what happened.
  • Uses statistical methods to summarize data.
• Diagnostic Analytics: Analyzes data to identify the causes of trends and correlations.
  • Examines why something happened.
  • Uses techniques like hypothesis testing, regression analysis, and correlation analysis.
• Predictive Analytics: Uses statistical models and machine learning to forecast future events.
  • Predicts what might happen.
  • Identifies potential opportunities and risks.
• Prescriptive Analytics: Recommends actions based on the results of predictive analytics.
  • Provides a plan of action.
  • Identifies the best course of action to achieve a desired outcome.

Data Analysis Techniques

• Quantitative Data Analysis: Analyzes numerical data using statistical methods.
  • Focuses on quantifiable data.
  • Uses mathematical calculations to make decisions.
• Qualitative Data Analysis: Analyzes non-numerical data using contextual insights.
  • Focuses on understanding the why behind user behavior.
  • Uses techniques like open-ended survey responses, interview clips, and behavioral observations.

Descriptive Statistics

• Summarize and describe data.
• Use statistical characteristics, charts, graphics, or tables.
• Focus on central tendency and dispersion.
• Includes measures like mean, median, mode, standard deviation, and variance.
• Used to describe the properties of a sample, but does not involve generalizing beyond the data.

Cross-Tabulation Analysis

• Analyzes categorical data.

• Compares results for one or more variables with the results of another.

• Presents data in a table format.

• Used to examine relationships within the data.

• Helps identify patterns and trends that may not be readily apparent.### Cross-Tabulation Analysis

• Useful when you have information that can be divided into categories or subgroups, such as age, gender, product type, or region.

• Examples of cross-tabulation analysis:

  • Newsletter signups by age
  • Preference by gender
  • Sales by region
  • Job seniority by education level
  • Product categories by payment method

Rule 1: Cross-Tabulation

• Cross-tabulation doesn't work as a statistical tool for numerical data, like if you wanted to sort through a list of heights or weights.
• You'd need to group this numerical data into categories for cross-tabulation to be effective.

Rule 2: Sample Size

• You must have enough data in your sample size for accurate cross-tab reports.
• The rule of thumb is that each entry in your data table should have a value of at least 5.

Mode, Median, and Average

• Mode: the most common answer in a data set, which means you use it to discover the most popular response for questions with numeric answer options.
• Median: reveals the middle of the road of your quantitative data by lining up all numeric values in ascending order and then looking at the data point in the middle.
• Average (mean): finding the average of a dataset is an essential quantitative data analysis method.
  • First, add all your quantitative data points, like numeric survey responses or daily sales revenue.
  • Then, divide the sum of your data points by the number of responses to get a single number representing the entire dataset.

Inferential Statistics

• A branch of statistics that uses various analytical tools to draw conclusions about the population from sample data.
• Inferential statistics is used to make inferences and draw conclusions; that is, to make valid generalizations from samples.
• Examples of inferential statistics:
  • Testing hypotheses and exploring relationships between variables
  • Correlation coefficients to explore the relationship between variables (e.g., whether there is a relationship between stress levels and academic results)

Key Differences: Descriptive vs. Inferential Statistics

• Descriptive statistics:
  • Purpose: describe and summarize data
  • Focuses on a dataset or subset of a population
  • Examples: mean, median, mode, standard deviation, range, frequency tables
  • Goal: summarize, organize, and present data
• Inferential statistics:
  • Purpose: make inferences and draw conclusions about a population
  • Focuses on a subset of the population (sample) to draw conclusions about the entire population
  • Examples: hypothesis testing, confidence intervals, regression analysis, ANOVA (analysis of variance), chi-square tests, t-tests, etc.
  • Goal: generalize findings to a larger population, make predictions, test hypotheses, evaluate relationships, and support decision-making