Quantitative Data Analysis

Training Objectives

• Introduce fundamental concepts and principles of quantitative data analysis
• Equip participants with practical skills in descriptive statistics for summarizing and describing data
• Show how to use correlational statistics to measure relationships between variables
• Provide knowledge of comparative statistics
• Show how to test hypotheses
• Train participants in regression analysis techniques to predict outcomes and model relationships
• Offer hands-on practice to reinforce learning and enhance proficiency in statistical techniques
• Empower participants to use methodologies effectively in research projects and academic endeavors

Contents

• Statistics and its branches
• Variables and their types
• Choosing the right statistical tool
• Introducing SPSS
• Descriptive statistics
• Correlational statistics
• Parametric comparative statistics
• Nonparametric comparative statistics
• Regression

What is Statistics?

• Statistics is the science of collecting, screening, analyzing, presenting, and interpreting data
• Data collection involves questionnaires, interviews, observations, tests, experiments, and/or registration
• Screening involves data selection, addition, and cleaning
• Analysis can be descriptive or inferential and can be correlational or comparative and parametric or nonparametric
• Data can be presented as textual, tabular, or graphical
• Interpretation involves determining guideline usage to interpret results

Branches of Statistics

• Descriptive Statistics
• Inferential Statistics
• Comparative statistics
• Correlational statistics
• Regression

Variables and Constants

• Variables are the characteristics that vary among members of a population or sample
• Constants are characteristics that do not vary

Variable Types According to Function

• Independent variables, also called predictors or variates, are measured and manipulated to test their effect on a given dependent variable(s)
• Dependent variables are the criterion variables which may or may not be affected by changes in the independent variable(s)

Variables according to continuity

• Continuous variables can take on infinite values between two values
• Weight (e. g. 52.6 kg), length (e. g. 16.33 cm), and grade (e. g. 81.5, 82.6) are all continuous variables
• Discrete variables can take countable values between any two values
• Number of siblings, Nara planted, and bicycles sold are discrete variables

Variables according to level of measurement

• Categorical variables, also called nominal variables, contain a finite number of categories that might not have logical order
• Sex (male or female), religion, schools (private or public), brand of cellphones, and types of computers are categorical variables
• Ordinal variables also have categories, but these can be ranked
• SES, academic rank, rank in a 1 km run test, and rank in NCAE are ordinal variables
• Interval numeric variables such as temperature in Celcius, allow the comparison of intervals and differences that mean something but its 'zero' point is not the real origin
• Ratio numeric variables such as temperature in Kelvin allow comparison of intervals or differences that mean something and its zero point is the real origin

Likert Scales

• Likert scale data are continuous
• Likert type data are ordinal
• Data Analysis procedures for Likert Type data are:
  • Central Tendency is median or mode
  • Variability is frequencies
  • Associations is Kendall tau B or C
  • Other Statistics include Chi-square
• Data Analysis procedures for Likert Scale data are:
  • Central Tendency is mean
  • Variability is standard deviation
  • Associations is Pearson's r
  • Other Statistics include ANOVA, t-test, regression

Choosing the Right Statistical Tool

• "What" questions are analyzed using descriptive statistics
• Categorical variables use frequency and percent
• Continuous use mean and SD
• Inferential statistical tests are used for hypothesis testing (yes-no questions)

Comparing Unpaired Groups

• When comparing unpaired groups with a continuous DV (parametric) with 2 groups the best statistical tool is an independent samples t-test
• When comparing unpaired groups with a categorical DV with 2 groups the best statistical tool is a chi-square test
• When comparing unpaired groups with a continuous DV (parametric) with more than 2 groups the best statistical tool is an ANOVA
• When comparing unpaired groups with a categorical DV with more than 2 groups the best statistical tool is a chi-square test

Comparing Paired Groups

• When comparing data with a 2 trials and continuous (parametric) variables the best tool is a Paired samples t-test
• When comparing data with over 2 trials and continuous (parametric) variables the best tool is a Repeated Measures ANOVA

Correlating Variables

• When correlating variables, with 2 continuous (Parametric) variables you would use Pearson-r
• When correlating variables, with 2 ordinal (Ranked numeric) variables you would use Spearman-rho/Kendall's tau-b
• When correlating variables, with categorical variables you would use Chi-square
• When correlating variables, with 2 variables with 1 covariate you would use Partial Correlation

Learn the core concepts of quantitative data analysis. This lesson covers descriptive, correlational, and comparative statistics. Also learn about regression analysis and hypothesis testing using SPSS for data analysis.