Here's the deal. These are two datasets that you are going to be using to do the graded exercise.
Here are descriptions of those datasets and the actual data.
# startle.txt
# These data simulate data that were collected by me during my disseration
# research at UCLA, 1980. Refs: King, W. B. (1980). Long-lasting effects of
# chronic administration of LSD on rat social and emotional behavior as an
# animal model for schizophrenia. Doctorial Dissertation, University of
# California, Los Angeles. King, Jr., W. & Ellison, G. (1989). Long-lasting
# alterations in behavior and brain neurochemistry following continuous
# low-level LSD adminstration. Pharmacology, Biochemistry, & Behavior, 33,
# 69-73. Three groups of rats were prepared. The control group received no
# LSD but received placebo injections and sham implants. The injection group
# received a total of 80 micrograms of LSD in 10 equal daily injections and
# sham implants. The minipump group received a total of 80 micrograms of LSD
# via a subcutaneously implanted osmotic minipump that continuously delivered
# the drug over a period of 10 days and placebo injections. After the ten days
# of drug administration, the animals were left undisturbed in their home cages
# for 30 days, except for routine cleaning and feeding. Behavioral testing
# began after that 30-day period, during which time the drug should have been
# completely cleared from the animals' bodies. A wide variety of behavioral
# tests were done. These data are from the startle test. The animal was placed
# in a small cage with a spring-loaded floor, then periodically blasted with
# a loud noise (120 dB if I recall). This caused the animals to startle, and
# the magnitude of the startle was measured by how much they jiggled the floor
# of the cage (transduced into a voltage). The numbers in the data represent
# startle magnitude in millivolts. The idea was that no one had ever been able
# to demonstrate long-lasting effects of LSD in animals because of the way
# the drug was administered (daily injections). We expected to see a long-
# lasting effect of the LSD in the minipump group. This is a wide format
# data frame. The same data occur in long format in startle.csv.
#
# Here's the odd thing. When the animals were randomized to groups, they were
# first matched for body weight to make sure that one group would not be lighter
# or heavier than the others. This is a typical procedure in drug experiments.
# However, this blocking variable is usually not included in the analysis, and
# data such as these are typically analyzed by one-way ANOVA. What is the
# disadvantage of that? We're about to find out!
#
control injection minipump
5.8 14.6 15.8
6.1 11.9 8.5
10.7 24.9 17.8
14.4 19.4 10.8
23.2 24.9 16.9
19.1 17.4 28.8
28.8 21.2 35.6
34.5 28.0 26.8
32.4 35.6 32.7
35.1 40.3 46.6
50.3 39.2 48.9
53.7 40.4 47.2
39.3 35.3 57.2
48.9 53.7 69.4
45.3 51.3 72.9
42.5 53.0 82.7
# chess.txt
# What is it that makes experts better than the rest of us at a task? Is it
# some general ability that they excel in, or is it a task-specific skill
# that they have developed through practice? In chess, expertise can be
# easily quantified and recognized, which makes it an ideal venue in which
# to study expertise. One of the first researchers to do this was a Dutch
# psychologist and chess master named de Groot (1946, trans. 1965). He
# found that chessmasters do not have better memories than anyone else,
# except when it comes to chess-related material such as board positions.
# His study was expanded upon by Chase & Simon (1973), Perception in chess,
# Cognitive Psychology, v.4, 55-81. They studied novice, average, and
# expert chess players, examining their ability to reproduce chess
# positions that they were able to examine only briefly. Two types of
# positions were used, positions that arose from a real game of chess,
# and positions that were produced by randomly placing pieces on the
# board. The measured variable was how well the subjects were able to
# recall the positions they saw (higher values mean more recall). The
# following data are not real but were reconstructed to display the same
# effects seen by Chase & Simon. The design is a 3x2 mixed factorial
# design with repeated measures on the position type variable. (This is
# not a faithful reproduction of Chase and Simon's experiment but displays
# some of the same effects.) Column 1 is a subject identifier which may be
# put into the row names. This is a wide format data frame with one subject
# per row. These same data occur in long format in chess.csv.
#
subject skill real random
N1 novice 36 58
N2 novice 27 26
N3 novice 57 61
N4 novice 32 34
N5 novice 37 29
N6 novice 59 57
N7 novice 32 45
N8 novice 65 63
N9 novice 29 32
A1 average 71 43
A2 average 39 66
A3 average 44 31
A4 average 77 46
A5 average 68 67
A6 average 69 59
A7 average 42 36
A8 average 40 36
A9 average 49 33
E1 expert 65 59
E2 expert 70 66
E3 expert 96 54
E4 expert 98 62
E5 expert 96 54
E6 expert 73 38
E7 expert 67 31
E8 expert 70 33
E9 expert 87 38
You can use RStudioCloud if you do not have access to an installed version of R.
The following commands would be used to fetch the datasets from the website. The first command clears your workspace, so make sure you're ready for that if you are copying and pasting.
rm(list=ls()) file1 = "http://ww2.coastal.edu/kingw/psyc480/data/startle.txt" startle = read.table(file=file1, header=T) file2 = "http://ww2.coastal.edu/kingw/psyc480/data/chess.txt" chess = read.table(file=file2, header=T, stringsAsFactors=T)
You will also need these two functions from the website.
source("http://ww2.coastal.edu/kingw/psyc480/functions/rmsaov.R")
source("http://ww2.coastal.edu/kingw/psyc480/functions/mixaov.R")
Your workspace should now look like this.
> ls() # six objects in your workspace [1] "chess" "file1" "file2" "mixaov" "rmsaov.test" [6] "startle"
Obviously the first dataset, startle.txt, is a single-factor treatment by blocks design. You will be using rmsaov.test() to get the ANOVA. The second dataset is a mixed factorial design. You will be using mixaov() to get the ANOVA. You should know how to do the following things.