LabWriteUp is a report writing application for students and researchers in science

 

LabWriteUp Home

New Project

Author Details

Open & Save Files

Design

Title

Aims & Hypothesis

Participants

Materials

Procedure

Results

Introduction

Discussion

Abstract

References

Build Report

APA Style Check

Export
to Word

Abbreviations

Quotations

Link words

 

Factorial Analysis of Variance (ANOVA)

Use the factorial ANOVA option to enter details of ANOVA tests that have more than two IVs. When this test is selected, you are presented with the list of variables defined in the design section. You can select two variables or more (hold down 'Ctrl' and click the variables). Once the variable has been selected, you can then click 'OK' to generate a brief description of the test.

Click the 'Next' button to continue. With a factorial ANOVA, there are several analyses to consider: the main effects of each of the independent variables and the interactions between pairs of them. LabWriteUp allows you to enter the data for each of these analyses independently.

If you want to produce a textual, graphical, or tabulated description of the summary data, click the 'Enter data for plot, table, or text'. However, due to the complexity when there are three IVs or more, tables can only be created for three IVs in LabWriteUp, and graphs can only be created for two IVs. If you want to create interaction plots with more than two levels, you can break the interaction into more than one plot (see later in this section).

A data grid appears with all cells of the interaction shown in rows:

In this example there are three IVs, each with two levels. Therefore there are 2 x 2 x 2 = 8 cells (shown as rows). If you wish to return to this data set later, you can save the data by selecting 'Save this data'.

To create graphs with three IVs. Suppose the three IVs are, anxiety group (high or low), item type (words or pictures), and valence (positive or negative). We could first plot the interaction for the high anxiety group and then plot the interaction for the low anxiety group. Essentially, we create an interaction between item type and valence (and enter the data for the high anxiety group) and then repeat this but for the low anxiety group.

We can then merge the two graphs with additional text, as created here:

 

Next >> Regression

 

Factorial ANOVA is a term used to describe complex ANOVAs, where there are three or more independent variables. The problem with factorial ANOVAs is that imteractions can be difficult to interpret. For this reason, ANOVAs should be 'theory driven', that is, we should be clear about the nature of the interaction we predict before wading through the results.