NCHRP 20-45
Scientific Approaches
to Transportation Research

Table of Contents

The following is a "hyperlinked" table of contents. By clicking on the desired outline heading, your browser will be directed to the appropriate section in the HTML documentation.

Vol. 1 Summary - Overview of Volume 1

Vol. 1 Chapter 1 - Introduction

Purpose and Scope

Basic Research, Applied Research, and Development

Transportation Research

Professional and Ethical Research Practices

Techniques and the Treatment of Experimental Data

Values in Science

Conflicts of Interest

Publication and Openness

Allocation of Credit

Authorship Practices

Acknowledgments

Error, Negligence, and Misconduct in Scientific Research

Individual Response to Violations of Ethical Standards

Effective Research

Vol. 1 Chapter 2 - Principles of Scientific Inquiry

The Scientific Method

Cause and Effect

Analysis and Synthesis

Induction, Deduction, and Abduction

Models and Mathematics

Necessary and Sufficient Conditions

Fallacies and Obstacles to Good Science

Vol. 1 Chapter 3 - The Research Process-Overview

Problem Statement Development

Project Selection

Requests for Qualifications or Proposals

Research Proposals

Review of Proposals and Contractor Selection

Research Work Plan

Execution of the Research

Implementation and Dissemination of Research Results

Recommended Practices

Evaluation of Research Projects

Miscellaneous Tools and Considerations

Client Involvement in the Research Process

Care and Nurturing of the Researchers

Vol. 1 Chapter 4 - Data Collection and Management

Questions Arising Before Data Collection

Questions Arising During Data Collection

Questions Arising after Data Collection

Questions Arising after Data Analysis

Questions Arising after the Study is Complete

Vol. 1 Chapter 5 - Reports and Presentations

Organization of Reports

Content of Reports

Uses and Misuses of Graphs

Copyright Issues

Vol. 1 Appendix A - Literature Searches

Vol. 1 Appendix B - The Research Process - Details and Examples

Vol. 1 Appendix C - Writing and Format of Reports

Vol. 1 Appendix D - Peer Review of Technical and Scientific Papers

Vol. 1 Appendix E - Checklist

Vol. 1 References

Vol. 2 Summary - Summary of Volume 2

Roadmap to Volume 2

Vol. 2 Chapter 1 - Identification of Empirical Setting

Purpose for Identifying the Empirical Setting

Inputs/Assumptions Assumed of the Investigator

Outputs/Products of this Chapter

Troubleshooting: Empirical Setting Frequently Asked Questions

References on Empirical Setting

Purpose of investigation

Predictive Research

Explanatory Research

Research for Quality Assurance and Control

Types of Research Investigations

Quasi-experiments

Observational Studies

Statistical Models and Modeling

Probabilistic Models

Decision-Making Models

Optimization Models

Dynamic Systems Models

Types of Variables Used in Models

Dependent and Independent Variables

Manipulated Variables

Randomization Variables

Exogenous and Endogenous Variables

Latent, Proxy, and Surrogate Variables

Indicator or Dummy Variables

Scales of Variable Measurement

Hypothesis Testing and Its Role in Statistical Model Development

Step 1: Generate the Null and Alternative Statistical Hypotheses from the Engineering Project under Investigation

Step 2: Select Test Statistic and Identify Assumptions of Statistical Distributions

Step 3: Formulate the Decision Rule and Set Alpha and Beta.

Step 4: Randomly Sample from the Population, Check the Statistical Assumptions, and Compute the Test Statistic.

Step 5: Apply the Decision Rule and Draw Conclusions

Misconception #1: Alpha is the Most Important Error Rate

Misconception #2: Hypothesis Test Results are Unconditional

Misconception #3: Hypothesis Test Conclusions are Correct

Misconception #4: Statistical Significance (Not Effect Size) Determines Practical Significance

Misconception #5: Non-Significant Hypothesis Test Results are Unimportant

Sample Size Determination

Model Validation

Internal Validity

External Validity

Troubleshooting: Empirical Setting Frequently Asked Questions

Vol. 2 Chapter 2 - Data Management, Analysis Tools, and Analysis Mechanics

Purpose of Data Management

The Data Collection, Handling, and Management Plan

Data storage: databases and data warehouses

Useful Data Handing Tools

Vol. 2 Chapter 3 - Identification of Appropriate Statistical Method

Purpose of this chapter

Basic Assumptions/Requirements of this chapter

Vol. 2 Chapter 4 - Continuous Dependent Variable Models

Purpose of Analysis of Variance:

Alternative Analysis of Variance Designs and Their Applications

Basic Assumptions/Requirements of Analysis of Variance:

Inputs for Analysis of Variance:

Outputs of Analysis of Variance:

General Analysis of Variance Methodology:

Interpretation of Analysis of Variance Output:

Troubleshooting: Analysis of Variance

Examples in Analysis of Variance:

Analysis of Variance References:

Analysis of Variance Methodology:

Interpretation of Analysis of Variance Output:

Troubleshooting: Analysis of Variance

Purpose of Linear Regression:

Basic Assumptions/Requirements of Linear Regression:

Alternative Regression Techniques and Their Applications

Inputs for Linear Regression:

Outputs of Linear Regression:

OLS Linear Regression Methodology:

Examples of Linear Regression:

Interpretation of Linear Regression Output:

Troubleshooting: Linear Regression

Linear Regression References:

Linear Regression Methodology:

Interpretation of Linear Regression Output:

Troubleshooting: Linear Regression

Purpose of Time Series Analysis:

Basic Assumptions/Requirements of Time Series Analysis:

Inputs for Time Series Analysis:

Outputs of Time Series Analysis:

Time Series Analysis Methodology:

Examples of Time Series Analysis:

Interpretation of Time Series Analysis Output:

Troubleshooting: Time Series Analysis

Time Series Analysis References:

Time Series Analysis Methodology:

Interpretation of Time Series Analysis Output:

TroubleShooting: Time Series Analysis

Purpose of Survival and Hazard Models:

Basic Assumptions/Requirements of Survival and Hazard Models:

Inputs for Survival and Hazard Models:

Outputs of Survival and Hazard Models:

Overview of Survival and Hazard Methodology:

Survival and Hazard Model References:

Vol. 2 Chapter 5 - Discrete Dependent Variable Models

Purpose of Logit, Nested Logit, and Probit Models:

Basic Assumptions/Requirements of Logit, Nested Logit, and Probit Models:

Inputs for Logit, Nested Logit, and Probit Models:

Outputs of Logit, Nested Logit, and Probit Models:

Logit, Nested Logit, and Probit Methodology:

Examples of Logit, Nested Logit, and Probit:

Interpretation of Logit, Nested Logit, and Probit:

Troubleshooting: Logit, Nested Logit, and Probit:

Logit, Nested Logit, and Probit References:

Logit, Nested Logit, and Probit Methodology:

Interpretation of Logit, Nested Logit, and Probit:

TroubleShooting: Logit, Nested Logit, and Probit

Purpose of Poisson and Negative Binomial Models:

Basic Assumptions/Requirements of Poisson and Negative Binomial Models:

Inputs for Poisson and Negative Binomial Models:

Outputs of Poisson and Negative Binomial Models:

Overview of Poisson and Negative Binomial Models:

Examples of Poisson and Negative Binomial Regression Models:

Poisson and Negative Binomial Model References:

Purpose of Logistic Regression Models:

Basic Assumptions/Requirements of Logistic Regression Models:

Inputs for Logistic Regression Models:

Outputs of Logistic Regression Models:

Overview of Logistic Regression Models:

Logistic Regression Examples:

Logistic Regression Model References:

Vol. 2 Chapter 6 - Nonparametric Methods

Purposes of Nonparametric Methods:

Basic Assumptions/Requirements; Nonparametric Vs. Parametric Methods:

Examples of Nonparametric Methods:

Nonparametric References:

Purpose of Graphical Methods:

Scatter Plots, Pairwise Scatter Plots, and Brush and Spin

Three Dimensional Graphics: Contours and Response Surfaces

Purpose of Descriptive Techniques:

Basic Descriptive Statistics of Data: Mean, Median, and Quartiles

Frequency Distributions, Variance, Standard Deviation, Histograms, and Boxplots

Purposes of Ranking and Counting Methods:

Ranking and Counting Methodology for Hypotheses About Medians

Ranking and Counting Methodology for Hypotheses About Medians (and Means):

Are the samples independent or dependent?

Hypotheses About Population Medians for Independent Samples

Wilcoxon Signed-Ranks Test - Hypothesis About Population Medians (Means) for a Single Independent Sample

ASSUMPTIONS OF WILCOXON SIGNED-RANKS TEST FOR SINGLE INDEPENDENT SAMPLE

INPUTS FOR WILCOXON SIGNED-RANKS TEST FOR SINGLE INDEPENDENT SAMPLE

HYPOTHESES OF WILCOXON SIGNED-RANKS TEST FOR SINGLE INDEPENDENT SAMPLE

TEST STATISTIC |D_i | OF WILCOXON SIGNED-RANKS TEST FOR SINGLE INDEPENDENT SAMPLE

INTREPRETATION OF OUTPUT (DECISION RULE) OF WILCOXON SIGNED-RANKS TEST FOR SINGLE INDEPENDENT SAMPLE

Mann-Whitney Test - Hypothesis About Population Means for Two Independent Samples

ASSUMPTIONS OF MANN-WHITNEY TEST FOR TWO INDEPENDENT SAMPLES

INPUTS FOR MANN-WHITNEY TEST FOR TWO INDEPENDENT SAMPLES

HYPOTHESES OF MANN-WHITNEY TEST FOR TWO INDEPENDENT SAMPLES

TEST STATISTIC FOR MANN-WHITNEY TEST FOR TWO INDEPENDENT SAMPLES

INTREPRETATION OF OUTPUT (DECISION RULE) OF MANN-WHITNEY TEST FOR TWO INDEPENDENT SAMPLES

Kruskal-Wallis One-Way Analysis Of Variance By Ranks Test For Several Independent Samples

ASSUMPTIONS OF KRUSKAL-WALLIS ONE-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL INDEPENDENT SAMPLES

INPUTS FOR KRUSKAL-WALLIS ONE-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL INDEPENDENT SAMPLES

HYPOTHESES OF KRUSKAL-WALLIS ONE-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL INDEPENDENT SAMPLES

TEST STATISTIC FOR KRUSKAL-WALLIS ONE-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL INDEPENDENT SAMPLES

INTREPRETATION OF OUTPUT (DECISION RULE) OF KRUSKAL-WALLIS ONE-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL INDEPENDENT SAMPLES

PAIRWISE COMPARISONS USING THE KRUSKAL-WALLIS ONE-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL INDEPENDENT SAMPLES

Wilcoxon Matched-Pairs Signed-Ranks Test For Two Dependent Samples

ASSUMPTIONS OF WILCOXON MATCHED-PAIRS SIGNED-RANKS TEST FOR TWO DEPENDENT SAMPLES

INPUTS FOR WILCOXON MATCHED-PAIRS SIGNED-RANKS TEST FOR TWO DEPENDENT SAMPLES

HYPOTHESES OF WILCOXON MATCHED-PAIRS SIGNED-RANKS TEST FOR TWO DEPENDENT SAMPLES

TEST STATISTIC |D_i | OF WILCOXON MATCHED-PAIRS SIGNED-RANKS TEST FOR TWO DEPENDENT SAMPLES

INTREPRETATION OF OUTPUT (DECISION RULE) OF WILCOXON MATCHED-PAIRS SIGNED-RANKS TEST FOR TWO DEPENDENT SAMPLES

Friedman Two-Way Analysis Of Variance By Ranks Test For Several Dependent Variables

ASSUMPTIONS OF FRIEDMAN TWO-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL DEPENDENT VARIABLES

INPUTS FOR FRIEDMAN TWO-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL DEPENDENT VARIABLES

HYPOTHESES OF FRIEDMAN TWO-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL DEPENDENT VARIABLES

TEST STATISTIC FOR FRIEDMAN TWO-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL DEPENDENT VARIABLES

INTREPRETATION OF OUTPUT (DECISION RULE) OF FRIEDMAN TWO-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL DEPENDENT VARIABLES

MULTIPLE COMPARISONS USING FRIEDMAN TWO-WAY ANALYSIS OF VARIANCE BY RANKS TEST FOR SEVERAL DEPENDENT VARIABLES

NONPARAMETRIC ANALYSIS OF BALANCED INCOMPLETE BLOCK DESIGNS

Purposes of Measures of Association and Tests for Independence:

Measures of Association and Tests for Independence Methodology:

Measures of Association and Tests for Independence Methodology:

What type of scale is used to measure the data?

Do you want a Measure of Association or a Test for Independence?

Measure of Association and Test for Independence for Nominal Scale Data

Cramer's Contingency Coefficient - Measure of Association for Nominal Scale Data

ASSUMPTIONS OF CRAMER'S CONTINGENCY COEFFICIENT

INPUTS FOR CRAMER'S CONTINGENCY COEFFICIENT

TEST STATISTIC (C) FOR CRAMER'S CONTINGENCY COEFFICIENT

INTREPRETATION OF OUTPUT OF CRAMER'S CONTINGENCY COEFFICIENT

Other Nonparametric Measures of Association for Nominal Scale Data

Chi-square Test for Independence (also called Cross Classification or Cross Tabulation) - Test for Independence for Nominal Scale Data

ASSUMPTIONS OF CHI-SQUARE TEST FOR INDEPENDENCE

INPUTS FOR CHI-SQUARE TEST FOR INDEPENDENCE

HYPOTHESES OF CHI-SQUARE TEST FOR INDEPENDENCE

TEST STATISTIC (T ) OF CHI-SQUARE TEST FOR INDEPENDENCE

INTEPRETATION OF OUTPUT (DECISION RULE) OF CHI-SQUARE TEST FOR INDEPENDENCE

APPROXIMATE VERSUS EXACT INFERENCES FOR CHI-SQUARE TEST FOR INDEPENDENCE AND CHI-SQUARE GOODNESS OF FIT TEST

Measure of Association and Test for Independence for Ordinal, Interval, and Ratio Scale Data

Kendall's Tau (t)-Measure of Association for Ordinal, Interval, and Ratio Scale Data

ASSUMPTIONS OF KENDALL'S TAU (t)

INPUTS FOR KENDALL'S TAU (t)

TEST STATISTIC (t) FOR KENDALL'S TAU

INTREPRETATION OF OUTPUT OF KENDALL'S TAU (t)

Other Nonparametric Measures of Association for Ordinal, Interval, and Ratio Scale Data

Hypothesis Test Using Kendall's Tau (t) - Test for Independence for Ordinal, Interval, and Ratio Scale Data

ASSUMPTIONS OF HYPOTHESIS TEST USING KENDALL'S TAU (t)

INPUTS FOR HYPOTHESIS TEST USING KENDALL'S TAU (t)

HYPOTHESES OF THE HYPOTHESIS TEST USING KENDALL'S TAU (t)

TEST STATISTIC (t) FOR HYPOTHESIS TEST USING KENDALL'S TAU

INTREPRETATION OF OUTPUT (DECISION RULE) OF HYPOTHESIS TEST USING KENDALL'S TAU (t)

Purposes of Goodness of Fit Methods:

Goodness-of-Fit Methodology:

Goodness-of-Fit (GOF) Methodology:

What type of scale is used to measure the data?

Are you testing one sample or multiple samples?

Goodness-of-Fit Tests for Nominal Scale Data - Chi-square Tests

Chi-square Goodness-of-Fit Test for Single Independent Sample

ASSUMPTIONS OF CHI-SQUARE GOF TEST FOR SINGLE INDEPENDENT SAMPLE

INPUTS FOR CHI-SQUARE GOF TEST FOR SINGLE INDEPENDENT SAMPLE

HYPOTHESES OF CHI-SQUARE GOF TEST FOR SINGLE INDEPENDENT SAMPLE

TEST STATISTIC (T ) OF CHI-SQUARE GOF TEST FOR SINGLE INDEPENDENT SAMPLE

INTREPRETATION OF OUTPUT (DECISION RULE) OF CHI-SQUARE GOF TEST FOR SINGLE INDEPENDENT SAMPLE

Chi-square Goodness-of-Fit Test for Two or More Independent Samples

ASSUMPTIONS OF CHI-SQUARE GOF TEST FOR TWO OR MORE INDEPENDENT SAMPLES

INPUTS FOR CHI-SQUARE GOF TEST FOR TWO OR MORE INDEPENDENT SAMPLES

HYPOTHESES OF CHI-SQUARE GOF TEST FOR TWO OR MORE INDEPENDENT SAMPLES

TEST STATISTIC (T ) OF CHI-SQUARE GOF TEST FOR TWO OR MORE INDEPENDENT SAMPLES

INTREPRETATION OF OUTPUT (DECISION RULE) OF CHI-SQUARE GOF TEST FOR TWO OR MORE INDEPENDENT SAMPLES

Trouble Shooting Chi-square Tests:

Goodness-of-Fit Tests for Ordinal, Interval, and Ratio Scale Data - Kolmogorov-Smirnov Type Tests

Kolmogorov Goodness-of-Fit Test for Single Independent Sample (Ordinal, Interval, and Ratio Scale Data)

ASSUMPTIONS OF KOLMOGOROV GOF TEST FOR SINGLE INDEPENDENT SAMPLE

INPUTS FOR KOLMOGOROV GOF TEST FOR SINGLE INDEPENDENT SAMPLE

HYPOTHESES OF KOLMOGOROV GOF TEST FOR SINGLE INDEPENDENT SAMPLE

TEST STATISTIC (T, T ⁺, T ^-) OF KOLMOGOROV GOF TEST FOR SINGLE INDEPENDENT SAMPLE

INTREPRETATION OF OUTPUT (DECISION RULE) OF KOLMOGOROV GOF TEST FOR SINGLE INDEPENDENT SAMPLE

Additional Nonparametric GOF Tests For Single Independent Sample (Ordinal, Interval, and Ratio Scale Data)

LILLIEFORS GOF TESTS FOR NORMAL AND EXPONENTIAL DISTRIBUTIONS FOR SINGLE INDEPENDENT SAMPLE

SHAPIRO-WILK GOF TEST FOR NORMAL DISTRIBUTION FOR SINGLE INDEPENDENT SAMPLE

Smirnov Goodness-of-Fit Test for Two Independent Samples (Ordinal, Interval, and Ratio Scale Data)

ASSUMPTIONS OF SMIRNOV GOF TEST FOR TWO INDEPENDENT SAMPLES

INPUTS FOR SMIRNOV GOF TEST FOR TWO INDEPENDENT SAMPLES

HYPOTHESES OF SMIRNOV GOF TEST FOR TWO INDEPENDENT SAMPLES

TEST STATISTIC (T, T ⁺, T^-) OF SMIRNOV GOF TEST FOR TWO INDEPENDENT SAMPLES

INTREPRETATION OF OUTPUT (DECISION RULE) OF SMIRNOV GOF TEST FOR TWO INDEPENDENT SAMPLES

Additional Nonparametric GOF Tests For Several Independent Samples (Ordinal, Interval, and Ratio Scale Data)

BIRNBAUM-HALL TEST FOR THREE INDEPENDENT SAMPLES

ONE-SIDED AND TWO-SIDED SMIRNOV TESTS FOR THREE OR MORE INDEPENDENT SAMPLES

Vol. 2 Appendix A - Alphabetical Glossary of Useful Statistical and Research Related Terms

Vol. 2 Appendix B - Useful Transformations

Purpose of Transformations

Precautions with Using Transformations of Variables

Rules of Thumb with Transformations

Examples of Common Transformations

Parabolic Transformations

Hyperbolic Transformations

Exponential Functions

Inverse Exponential Functions

Power Functions

Vol. 2 Appendix C - Statistical Tables

Vol. 2 Appendix D - References