UNIVERSITY OF CAGAYAN VALLEY

(Formerly Cagayan Colleges Tuguegarao)

Tuguegarao City, Cagayan, Philippines
SCHOOL OF LIBERAL ARTS AND TEACHER EDUCATION
FACTORIAL ANOVA

A factorial ANOVA is an Analysis of Variance test with more than one independent variable, or “factor“. It can
also refer to more than one Level of Independent Variable. For example, an experiment with a treatment group
and a control group has one factor (the treatment) but two levels (the treatment and the control). The terms
“two-way” and “three-way” refer to the number of factors or the number of levels in your test. Four-way
ANOVA and above are rarely used because the results of the test are complex and difficult to interpret.

A two-way ANOVA has two factors (independent variables) and one dependent variable. For example, time
spent studying and prior knowledge are factors that affect how well you do on a test.
A three-way ANOVA has three factors (independent variables) and one dependent variable. For example, time
spent studying, prior knowledge, and hours of sleep are factors that affect how well you do on a test
Factorial ANOVA is an efficient way of conducting a test. Instead of performing a series of experiments where
you test one independent variable against one dependent variable, you can test all independent variables at the
same time.

The factorial univariate ANOVA is an extension of the one-way ANOVA in that it involves the analysis of two
or more independent variables. It is used in experimental designs in which every level of every factor is paired
with every level of every other factor.

It allows the researcher to assess the effects of each independent variable separately, as well as the joint effect
or interaction of variables. Factorial designs are labelled either by the number of factors involved or in terms of
the number of levels of each factor. Thus, a factorial design with two independent variables (e.g., gender and
ethnicity) and with two levels for each independent variable (male/female; Australian/Chinese) is called either a
two-way factorial or a 2 × 2 factorial.

CHECKLIST OF REQUIREMENTS
 In any one analysis, there must be two or more independent variables (due to the complexity in
interpreting higher order interactions, most factorial designs are limited to three or four independent
variables or factors).
 There can be two or more levels for each independent variable.
 There must be only one dependent variable.

ASSUMPTIONS
 The populations from which the samples were taken are normally distributed (Test of Normality).
 Homogeneity of variance.
 The observations are all independent of one another.

Test of Normality -Shapiro-Wilk and Kolmogorov Test of Normality

The above table presents the results from two well-known tests of normality, namely the Kolmogorov-Smirnov
Test and the Shapiro-Wilk Test. The Shapiro-Wilk Test is more appropriate for small sample sizes (< 50
samples), but can also handle sample sizes as large as 2000. For this reason, we will use the Shapiro-Wilk test
as our numerical means of assessing normality.
We can see from the above table that for the "Beginner", "Intermediate" and "Advanced" Course Group the
dependent variable, "Time", was normally distributed. How do we know this?

If the Sig.  value of the Shapiro-Wilk Test is greater than 0.05, the data is normal. If it is below 0.05, the data
significantly deviate from a normal distribution.
 UNIVERSITY OF CAGAYAN VALLEY
(Formerly Cagayan Colleges Tuguegarao)
Tuguegarao City, Cagayan, Philippines
SCHOOL OF LIBERAL ARTS AND TEACHER EDUCATION
BOX PLOTS

Box plots (also called box-and-whisker plots or box-whisker plots) give a good graphical image of the
concentration of the data. They also show how far the extreme values are from most of the data. A box plot is
constructed from five values: the minimum value, the first quartile, the median, the third quartile, and the
maximum value. We use these values to compare how close other data values are to them.

To construct a box plot, use a horizontal or vertical number line and a rectangular box. The smallest and largest
data values label the endpoints of the axis. The first quartile marks one end of the box and the third quartile
marks the other end of the box. Approximately the middle 50 percent of the data fall inside the box. The
“whiskers” extend from the ends of the box to the smallest and largest data values. The median or second
quartile can be between the first and third quartiles, or it can be one, or the other, or both. The box plot gives a
good, quick picture of the data.

FACTORIAL ANOVA Example: Two-way Factorial (2 × 2 Factorial)

A researcher is interested in determining the effects of two learning strategies (A and B) on the memorization of
a hard vs. an easy list of syllables. The factorial combination of these two independent variables (2 × 2) yields
four experimental conditions: Strategy A–Easy List, Strategy A–Hard List, Strategy B–Easy List, and Strategy
B–Hard List. A total of 24 subjects are randomly assigned to the four experimental conditions. The researcher
recorded the total number of errors made by each subject.

Strategy A Strategy B
Easy List S1 6 S13 20
S2 13 S14 18
S3 11 S15 14
S4 8 S16 14
S5 9 S17 12
S6 5 S18 16

Hard List S7 15 S19 16

S8 17 S20 13
S9 23 S21 15
S10 21 S22 20
S11 22 S23 11
S12 20 S24 12
STEPS: Input in SPSS

Variable View
 UNIVERSITY OF CAGAYAN VALLEY
(Formerly Cagayan Colleges Tuguegarao)
Tuguegarao City, Cagayan, Philippines
SCHOOL OF LIBERAL ARTS AND TEACHER EDUCATION
Data View
 UNIVERSITY OF CAGAYAN VALLEY
(Formerly Cagayan Colleges Tuguegarao)
Tuguegarao City, Cagayan, Philippines
SCHOOL OF LIBERAL ARTS AND TEACHER EDUCATION
 UNIVERSITY OF CAGAYAN VALLEY
(Formerly Cagayan Colleges Tuguegarao)
Tuguegarao City, Cagayan, Philippines
SCHOOL OF LIBERAL ARTS AND TEACHER EDUCATION
 UNIVERSITY OF CAGAYAN VALLEY
(Formerly Cagayan Colleges Tuguegarao)
Tuguegarao City, Cagayan, Philippines
SCHOOL OF LIBERAL ARTS AND TEACHER EDUCATION
 UNIVERSITY OF CAGAYAN VALLEY
(Formerly Cagayan Colleges Tuguegarao)
Tuguegarao City, Cagayan, Philippines
SCHOOL OF LIBERAL ARTS AND TEACHER EDUCATION
 UNIVERSITY OF CAGAYAN VALLEY
(Formerly Cagayan Colleges Tuguegarao)
Tuguegarao City, Cagayan, Philippines
SCHOOL OF LIBERAL ARTS AND TEACHER EDUCATION

Figure 1

TEST OF NORMALITY

Steps: ANALYZE - DESCRIPTIVE STATISTICS- EXPLORE

CLICKNormality Plots and Stem-And-Leaf Then CLICK Ok

 UNIVERSITY OF CAGAYAN VALLEY
(Formerly Cagayan Colleges Tuguegarao)
Tuguegarao City, Cagayan, Philippines
SCHOOL OF LIBERAL ARTS AND TEACHER EDUCATION

OUTPUT:

Tests of Normality
strategy Kolmogorov-Smirnova Shapiro-Wilk

Statistic df Sig. Statistic df Sig.

*
Strategy A .151 12 .200 .930 12 .381
errors
Strategy B .140 12 .200* .930 12 .378

*. This is a lower bound of the true significance.

a. Lilliefors Significance Correction

RESULTS AND INTERPRETATION

Main Effect

 The main effect of STRATEGY is not significant, F(1,20) = 0.53, p > .05. From the estimated marginal
means, the difference in the number of errors made by the strategy A group (M = 14.167) is not
significantly different from the number of errors made by the strategy B group (M = 15.083)
 The main effect of LIST is significant, F(1,20) = 15.31, p < .05. From the Estimated Marginal Means, it
can be seen that the subjects made significantly more errors in the hard list (M = 17.08) than in the easy
list (M = 12.16) (collapsing across the two STRATEGY levels).

Interaction Effect

 The STRATEGY*LIST interaction is significant, F(1,20) = 23.43, p < .001. To interpret the interaction,
the task is made easier by graphing the STRATEGY*LIST estimated marginal means as shown in
Figure 1.
 UNIVERSITY OF CAGAYAN VALLEY
(Formerly Cagayan Colleges Tuguegarao)
Tuguegarao City, Cagayan, Philippines
SCHOOL OF LIBERAL ARTS AND TEACHER EDUCATION

Figure 1

Interaction Effect

From Figure 1, it can be seen that the effect of learning strategy on the number of errors made is dependent on
the difficulty of the list learned. Under strategy A, subjects made more errors in the hard list than in the easy
list, but under strategy B, the effect is opposite, with subjects making more errors in the easy list than in the
hard list.

Post Hoc Test for Simple Effects

 The significant interaction effect indicates that the effect of one independent variable on the dependent
variable is dependent on the second independent variable, i.e., the four experimental conditions
(Strategy A–Easy List, Strategy A–Hard List, Strategy B–Easy List, and Strategy B–Hard List) differ
significantly in affecting the number of errors made.
 However, the interaction effect does not indicate where the differences are, i.e., between which
experimental conditions. To identify specific differences, Post Hoc comparisons can be used to “tease
apart” the interaction. This is equivalent to the test for simple effects, i.e., the effect of one factor (IV1 )
at one level of the other factor (IV2 ).
 Unfortunately, in a factorial design such as this, post hoc comparisons between the four experimental
conditions cannot be directly conducted within the GLM Univariate program. Rather, post hoc
comparisons can only be executed through the ONEWAY ANOVA analysis. This procedure requires
some data manipulation to convert the four experimental conditions into four levels of the same
grouping variable. These four levels can then be compared using the Scheffé post hoc test via the
ONEWAY ANOVA analysis.

You can also watch this video to further deepen your understanding of the above topic.
https://www.youtube.com/watch?v=5yDoxhcqkcc

Watch the attached video to guide you on how you How to Use SPSS in Data Analysis.
https://www.youtube.com/watch?v=ulk_JWckJ78
https://www.youtube.com/watch?v=DcS6_I63PHs