# Module 3: Research Lab2- ANCOVA

## Module 3: Research Lab 2 – ANCOVA

Start Assignment

• Points 75

### Purpose

This assignment is intended to help you learn to do the following:

• Assess the assumptions of ANCOVA.
• Conduct ANCOVA analysis.
• Interpret the results of ANCOVA analysis.

### Preparation

ANCOVA examines the influence of an independent variable on a dependent variable while removing the effect of the covariate factor. ANCOVA first conducts a regression of the independent variable (i.e., the covariate) on the dependent variable.

When used as an extension of multiple regression, ANCOVA can test all of the regression lines to see which have different Y intercepts as long as the slopes for all lines are equal.

Like regression analysis, ANCOVA enables you to look at how an independent variable acts on a dependent variable. ANCOVA removes any effect of covariates, which are variables you donâ€™t want to study. For example, you might want to study how different levels of teaching skills affect student performance in math. It may not be possible to randomly assign students to classrooms. Youâ€™ll need to account for systematic differences between the students in different classes (e.g. different initial levels of math skills between gifted and mainstream students).

Before you work on this assignment, complete the assigned readings listed in the Module 3 Preparation page.

### Overview

A consultant was hired by a large corporation to assess the performance of its branch offices located in Massachusetts, Texas, and Washington states regarding the number of projects completed by the offices in the last 24 months. Each branch office has 12 teams and projects are assigned as they become available. Assume that the IV is the branch_office coded as 1, 2, and 3 standing for Massachusetts, Texas, and Washington states, respectively. The DV is project_done and represents volume of projects completed by a team over the last 24 months. It has also become apparent to the consultant that there is another variable that could affect the number of projects completed and it might be pertinent to account for it. The thinking is that projects generally vary in their level of complexity and it is expected that more complex projects would take longer time that less complex projects. It is therefore more likely that holding all else constant, fewer projects would be completed if those projects were complex and more projects would be completed if those projects were simpler. It is clear therefore that just focusing on the number of projects completed alone without considering their complexity would not fully give an accurate rate of completing projects in the branch offices.

Based on this information, the consultant added project complexity as a variable to be considered in assessing the performance of branch offices. Project complexity is treated as a covariate in this scenario and will be named complexity_cv for purposes of analysis to be conducted.

### Action Items

Consider the following dataset: Ancova_projects_completed.xlsx

1. Screen the dataset for accuracy.
2. Run the dataset to check for assumptions of ANCOVA, in particular evaluate linearity of regression and homogeneity of regression assumptions. Was any of these assumptions violated? What will you recommend if any of these assumptions are violated?
3. Do the offices differ in the number of projects completed over a two-year period when project complexity is not considered?
4. Based on the output generated, is the global or omnibus test significant?
5. Do the offices differ in the number of projects completed over a two-year period when project complexity considered? Which office(s) if any completes least projects?
6. Write one to two paragraphs of how you will report this results in finding section of a dissertation or journal article.
7. What are the main reasons for using covariance analysis in a randomized study?

## Rubric

Module 3: Research Lab 2 – ANCOVA