Module 1 and Lecture 1

Module 1: Collecting Data

Topics Overview

  • Topic 1: The Structure of Data

  • Topic 2: Sampling from a Population

  • Topic 3: Experiments and Observational Studies

  • Reference: Covers Chapter 1 in Lock5 Text

Understanding Data

  • Definition of Data: Data are defined as observations gathered for analysis, which may also be described as numbers with a context. It is important to note that data do not have to be numerical.

  • Focus on Methodology: The course emphasizes the appropriate methods of data collection and organization.

The Language of Data

  • Dataset: A dataset is defined as a collection of values for one or more variables that record or measure information for each case.

    • Cases: These are the subjects or objects about which information is obtained and may also be referred to as subjects, units, or individuals. In a data table, each case is represented as a row.

    • Variables: Any characteristic that is recorded for each case, typically corresponding to the columns in a data table.

Characteristics of Cases and Variables

  • Cases: Objects described by a dataset such as individuals or any item from which data is gathered.

  • Variables: Characteristics of a case that can take different values across different cases.

  • Values of a Variable: These are the possible observations linked to the variable. Data refer to the observed values describing each case.

Example Scenario: College Applications

  • Referring to the students’ recent experience of applying for colleges as a case study.

  • Identified Cases: Various colleges or universities applied to by students.

    • Examples of cases include: GMU, UVA, Maryland, NYU, UCLA, Iowa State, Miami.

Variables in College Examples

  • Cost: Tuition cost of each college.

  • Location: Geographic location of each college.

  • Size: The size of the college, usually indicated by the number of students enrolled.

  • Football Team: Presence or absence of a football team.

Specific Data Entry for College Variables

  • Cost for Each College:

    • GMU: $15,000

    • UVA: $20,000

    • Maryland: $31,000

    • NYU: $60,000

    • UCLA: $55,000

    • Iowa State: $29,000

    • Miami: $47,000

  • Location (by residency):

    • GMU: In-State

    • UVA: In-State

    • Maryland: Out of State

    • NYU: Out of State

    • UCLA: Out of State

    • Iowa State: Out of State

    • Miami: Out of State

Types of Variables

  • Categorical Variables: These variables divide cases into groups, allowing each case to belong to exactly one category. They describe attributes of cases and are also known as qualitative data. Examples include:

    • Cost category (in-state vs. out-of-state)

    • Location type (urban, suburban, rural)

    • Size category (small, midsize, big)

    • Football team presence (yes or no)

  • Quantitative Variables: These variables measure or record numerical quantities for each case, allowing numerical operations such as adding and averaging. Also known as numerical data, examples include:

    • Cost (in dollars)

    • Distance (in miles)

    • Size (total number of students)

Examining Variable Relationships

  • The course poses critical questions about potential relationships between variables:

    • Does the cost of college correlate with a student's residency status?

    • Does the presence of a football team depend on the size of the college?

    • Is there a relationship between a college's size and its cost?

Types of Variables in Studies

  • When two variables are involved in a study, they can be classified as:

    • Explanatory Variable (Independent or Predictor Variable): The variable believed to “explain” the change in another variable. This is often manipulated by researchers.

    • Response Variable (Dependent Variable): The variable that is believed to be impacted by the explanatory variable, this variable “responds” to changes in the explanatory variable.

Identifying Variables in the College Example

  • Scenario Analysis:

    • Cost influenced by in-state residency status:

    • Explanatory: In-state or out-of-state status

    • Response: Cost

    • Influence of college size on football team presence:

    • Explanatory: Size of the college

    • Response: Existence of a football team

    • Relationship between size and cost:

    • No clear distinction - could be interpreted in either context.

Making Comparisons in Data Analysis

  • The objectives when making comparisons between variables include:

    • Identifying associations between the two variables.

    • Determining causation: Can conclusions about one variable affecting another be drawn based on the data sampled?

  • The answers to these inquiries depend significantly on how data are collected and the sampling methods employed.