PS-L2-Research-Variables

Lesson Overview

• Lesson 2 focuses on Research and covers the concept of Variables.

Research Variables and Measurement

• Research Variables: Factors that can be manipulated and measured in scientific experiments or research processes.

  • Characteristics/attributes that can have different values for different subjects.

  • Examples include age, educational qualifications, gender, and civil status.

• Measurement: The process of determining the value or label of a variable for a specific individual or object.

Types of Variables

Discrete and Continuous Variables

• Discrete Variables: can assume a finite or countably infinite number of values. Measured by counting.

  • Examples: Students, professors, children.

• Continuous Variables: Cannot be counted, they can take any value within a range.

  • Examples: Intelligence, beauty, effectiveness.

Qualitative and Quantitative Variables

Qualitative Variables

• Variables yielding categorical responses.

  • Examples: Occupation, gender, religious affiliation.

Quantitative Variables

• Variables that represent a numerical amount or quantity.

  • Examples: Height, salary, number of children.

Dependent and Independent Variables

• Independent Variables: Controlled or manipulated by the researcher.

• Dependent Variables: Measured based on the effects of the independent variables.

  • Example: Predictive validity of entrance requirements for freshman students; independent variables could be exam scores and dependent variable could be student performance.

Variable Distribution

Univariable, Bivariable, and Multivariable Distribution

• Univariable Distribution: Involves one variable only.

  • Example: Age of Grade 7 pupils.

• Bivariable Distribution: Involves two variables.

  • Example: Ice cream sales vs. temperature.

Bivariable Example

• Data shows correlation: As temperature rises, ice cream sales increase.

  • Example data:

    • Temperature: 14.2°C | Sales: Php 215

    • Temperature: 25.1°C | Sales: Php 614

Multivariable Distribution

• Involves three or more variables.

  • Example: Tracking college enrollment by program, year level, and gender.

Levels of Measurement

Nominal Scale

• No numerical value; categorical.

  • Examples: Gender, employment status.

Ordinal Scale

• Classes variables and ranks them.

  • Examples: Quality ratings (Outstanding to Poor).

Interval and Ratio Scales

Interval Scale

• Has nominal and ordinal characteristics with predetermined intervals but no true zero.

Ratio Scale

• Highest level of measurement with a true zero point.

  • Examples: Height, weight, time.

Population and Sample

• Population: Total group of interest.

• Sample: Subset of the population for data collection.

Steps in Determining Sample Size

Stage 1

• Identify population for data gathering.

Stage 2

• Determine the type of sample to be selected.

Stage 3

• Use Slovin’s formula to calculate sample size:

  • 𝑛 = 𝑁 / (1 + 𝑁𝑒²)

Sample Size Calculation Example

• Example: Finding sample size for reading deficiencies in 5,000 students with a 5% margin of error.

Representing Population

• Identified that testing 370 students provides representative results for the population of 5,000.

Parameters and Statistics

• Parameters: Population measures, symbolized by μ.

• Statistics: Measures computed about a sample, estimate of population characteristics.

Sampling Methods

Probability Sampling

• Each unit has a known nonzero probability of inclusion. Includes various techniques.

Non-Probability Sampling

• Individual units do not have specified selection probabilities.

  • Includes purposive, convenience, quota, and snowball sampling.

Probability Sampling Techniques

Simple Random Sampling

• Equal chance for each member, chosen randomly.

• Example: Fishbowl method.

Stratified Random Sampling

• Samples selected from various population groups (strata).

• Ensures more accurate representation.

Systematic Random Sampling

• Every kth member from a list is selected.

• Formula: K = N/n, where K is sampling interval, N is population, n is desired sample size.

Clustering and Multi-stage Sampling

Cluster Sampling

• Groups/population divided into clusters for sampling.

Multi-stage Sampling

• Larger populations sampled in stages, reducing resource needs.

Non-Probability Sampling Techniques

Purposive Sampling

• Participants selected based on judgment and relevance to the study.

Convenience Sampling

• Participants selected based on availability.

Additional Non-Probability Techniques

Quota Sampling

• Focus on obtaining desired sample size irrespective of selection method.

Snowball Sampling

• Initial respondents help identify further subjects within networks.

Research Design Overview

• Research design includes plans for selecting respondents and data gathering.

Types of Research Design

1. Action Research

2. Descriptive Research

3. Explanatory Research

4. Exploratory Research

5. Correlational Research

6. Evaluation Research

7. Policy Research

8. Ex-post Facto Research

9. Historical Research

10. Ethnographic Research

11. Phenomenological Research

DESCRIPTION

• Action Research: Focuses on solving an immediate problem while reflecting on the process and outcomes.

• Descriptive Research: Aims to describe characteristics of a population or phenomenon being studied. It does not test hypotheses but provides insights.

• Explanatory Research: Seeks to explain why or how phenomena occur. It often relies on causal analysis to understand relationships between variables.

• Exploratory Research: Conducted to clarify ambiguous problems or explore new areas where little is known. It generates insights and hypotheses rather than tests them.

• Correlational Research: Examines the relationship between two or more variables to identify associations without implying causation.

• Evaluation Research: Assesses the effectiveness of programs, policies, and practices, often using pre-defined criteria for success.

• Policy Research: Analyzes and evaluates the development and impact of policies. It provides evidence-based recommendations.

• Ex-post Facto Research: Involves looking back at existing data or events to infer relationships when controlled experiments are not possible.

• Historical Research: Explores past events to understand their meaning and significance. It typically relies on primary and secondary sources.

• Ethnographic Research: Involves immersive observation and direct interaction with a specific cultural or social group to understand their behaviors and perspectives.

• Phenomenological Research: Focuses on experiences and perceptions of individuals, seeking to understand the essence of those experiences.