bruceR (version 0.6.0)

MANOVA: Multi-factor ANOVA.

Description

Easily perform multi-factor ANOVA (between-subjects, within-subjects, and mixed design).

Usage

MANOVA(
  data,
  subID = NULL,
  dv = NULL,
  dvs = NULL,
  dvs.pattern = "",
  between = NULL,
  within = NULL,
  covariate = NULL,
  sph.correction = "none",
  which.observed = NULL,
  nsmall = 2
)

Arguments

data

Data frame. Both long-format and wide-format can be used.

• If you input long-format data, please also set subID.

• If you input wide-format data (i.e., one subject occupies one row, and repeated measures occupy multiple columns), the function can automatically transform the data into long-format.

subID

Subject ID.

• If you input long-format data, you should set the subject ID.

• If you input wide-format data, no need to set this parameter.

dv

Variable name of dependent variable.

• If you input long-format data, then dv is the outcome variable.

• If you input wide-format data, then dv can only be used for complete between-subjects design. For designs with any repeated measures, please use dvs and dvs.pattern.

dvs

[only for "wide-format" data and designs with repeated measures]

Variable names of repeated measures.

• You can use ":" to specify a range of variables: e.g., "A1B1:A2B3" (similar to the SPSS syntax "TO"; the variables should be put in order)

• You can also use a character vector to specify variable names: e.g., c("Cond1", "Cond2", "Cond3")

dvs.pattern

[only for "wide-format" data and designs with repeated measures]

If you set dvs, you must also set the pattern of variable names by using regular expressions.

Examples:

• "Cond(.)" can extract levels from "Cond1", "Cond2", "Cond3", ...

  You can rename the factor name by using within: e.g., within="Condition"

• "X(..)Y(..)" can extract levels from "X01Y01", "X02Y02", "XaaYbc", ...

• "X(.+)Y(.+)" can extract levels from "X1Y1", "XaYb", "XaY002", ...

Tips on regular expression:

• "(.)" extracts any single character (can be number, letter, or other symbols)

• "(.+)" extracts >= 1 character(s)

• "(.*)" extracts >= 0 character(s)

• "([0-9])" extracts any single number

• "([a-z])" extracts any single letter

• each pair of "()" extracts levels for each factor

between

Between-subjects factors. Character string (e.g., "A") or vector (e.g., c("A", "B")). Default is NULL.

within

Within-subjects factors. Character string (e.g., "A") or vector (e.g., c("A", "B")). Default is NULL.

covariate

Covariates (if necessary). Character string or vector. Default is NULL.

sph.correction

[only effective for repeated measures with >= 3 levels]

Sphericity correction method to adjust the degrees of freedom (df) when the sphericity assumption is violated. Default is "none". If Mauchly's test of sphericity is significant, you may set it to "GG" (Greenhouse-Geisser) or "HF" (Huynh-Feldt).

which.observed

[only effective for computing generalized \(\eta^2\)]

Factors that are observed or measured (e.g., gender, age group, measured covariates) but not experimentally manipulated. Default is NULL. The generalized \(\eta^2\) requires correct specification of the observed (vs. manipulated) variables. (If all the variables in between and within are set to observed, then generalized \(\eta^2\) will be equal to \(\eta^2\).)

nsmall

Number of decimal places of output. Default is 2.

Value

A result object returned by afex::aov_ez.

Details

This function is based on and extends the afex::aov_ez function in the R package afex. You only need to specify the data, dependent variable(s), and factors (between-subjects and/or within-subjects). Then, almost all the outputs you need will be displayed in an elegant manner, including effect sizes (partial \(\eta^2\)) and their confidence intervals (CIs). 90% CIs for partial \(\eta^2\) are reported, following the suggestion by Steiger (2004).

References

Olejnik, S., & Algina, J. (2003). Generalized eta and omega squared statistics: Measures of effect size for some common research designs. Psychological Methods, 8(4), 434-447. 10.1037/1082-989X.8.4.434

Steiger, J. H. (2004). Beyond the F test: Effect size confidence intervals and tests of close fit in the analysis of variance and contrast analysis. Psychological Methods, 9(2), 164-182. 10.1037/1082-989X.9.2.164

See Also

EMMEANS, bruceR-demodata

Examples

# NOT RUN {
#### Between-Subjects Design ####

between.1
MANOVA(data=between.1, dv="SCORE", between="A")

between.2
MANOVA(data=between.2, dv="SCORE", between=c("A", "B"))

between.3
MANOVA(data=between.3, dv="SCORE", between=c("A", "B", "C"))


#### Within-Subjects Design ####

within.1
MANOVA(data=within.1, dvs="A1:A4", dvs.pattern="A(.)",
       within="A")
## the same:
MANOVA(data=within.1, dvs=c("A1", "A2", "A3", "A4"), dvs.pattern="A(.)",
       within="MyFactor")  # renamed the within-subjects factor

within.2
MANOVA(data=within.2, dvs="A1B1:A2B3", dvs.pattern="A(.)B(.)",
       within=c("A", "B"))

within.3
MANOVA(data=within.3, dvs="A1B1C1:A2B2C2", dvs.pattern="A(.)B(.)C(.)",
       within=c("A", "B", "C"))


#### Mixed Design ####

mixed.2_1b1w
MANOVA(data=mixed.2_1b1w, dvs="B1:B3", dvs.pattern="B(.)",
       between="A", within="B")
MANOVA(data=mixed.2_1b1w, dvs="B1:B3", dvs.pattern="B(.)",
       between="A", within="B", sph.correction="GG")

mixed.3_1b2w
MANOVA(data=mixed.3_1b2w, dvs="B1C1:B2C2", dvs.pattern="B(.)C(.)",
       between="A", within=c("B", "C"))

mixed.3_2b1w
MANOVA(data=mixed.3_2b1w, dvs="B1:B2", dvs.pattern="B(.)",
       between=c("A", "C"), within="B")


#### Other Examples ####
data.new=mixed.3_1b2w
names(data.new)=c("Group", "Cond_01", "Cond_02", "Cond_03", "Cond_04")
MANOVA(data=data.new, dvs="Cond_01:Cond_04", dvs.pattern="Cond_(..)",
       between="Group", within="Condition")  # renamed the within-subjects factor

?afex::obk.long
MANOVA(data=afex::obk.long, subID="id", dv="value",
       between=c("treatment", "gender"), within=c("phase", "hour"), cov="age",
       sph.correction="GG")
# }