Data sets that have been discussed in class or are used in homework are:

• Fishoil.txt Data file for HW 1
• Tomato.txt Tomato data file for SAS intro workshop.
• Insect.txt Insect data file for SAS intro workshop.
• Insect.xls Excel worksheet version of insect data set.
• Food.txt Data file for food palatibility study. First column is name of the protein supplement (c, l, s). Second column is a preference score (-3 to 3) where 3 = most palatible.
• Salinity.txt Data file for barley salinity study. Used to illustrate variance component calculations
• cotton.txt Data on contamination in seed lots for HW 4. Columns are lot sample contamination.
• porosity.txt Porosity data from HW 5. Columns are field id, section id, porosity value.
• poro2.txt Corrected porosity data for HW 5. Columns are field id, section id, porosity value.
• blockex.txt Data for blockex.sas. Example from Mead, section 5.3, arranged for a paired t-test, i.e. one row per block (= per pair of e.u.'s)
• range.txt Response of blue grama grass to rangeland fertilization. Variables are the treatment, the block number, the response: plant phosphate concentration, and a treatment code (a - e) that you can ignore or use instead of the treatment name.
• metab.txt Metabolite data set for Midterm I. Columns are the plant id, the extract id, the measurement id, and the measured concentration.
• fullmoon.txt Data on admissions to mental health clinic for Midterm I. Columns are the month, the phase of the moon (B: immediately before full moon, D: during full moon, and A: immediately after full moon), and the admissions rate for that period (# patients / day).
• soytaste.txt Tofu taste scores for tofu tasting problem on HW 6. Columns are the Person, Tofu type, and beaniness score (0 = not beany, 15 = very beany).
• plankton.txt Data for plankton abundance problem on HW 6. Columns are tow (i.e. the block), species, and abundance.
• millet.txt Millet row spacing study using a latin square design for HW 6. Columns are the row, column, treatment (spacing), and yield (response).
• millet2.txt Millet row spacing study using two 5 x 5 latin square design for HW 6.. Columns are the square, row, column, treatment (spacing), and yield (response).
• multblock.txt Data for example of multiplicative block effects.
• brome.txt Data for example of analysis of a Latin Square. The row blocking variable is proximity to freeway; the column blocking variability is proximity to a stream. Columns in the data file are the row, column, treatment, and a measure of the abundance of the rare species.
• brome2.txt Data for illustration of multiple latin squares. Made up data based on brome grass study, 3 squares, each 4x4.
• Food2.txt Data file for entire food palatability study. First column is the treatment code (1 through 6), second column is the sex of the participant, third colume is the type of the protein supplement (c, l, s). Last column is a preference score (-3 to 3) where 3 = most palatable.
• carrot.txt Carrot experiment to look at effects of seed stock and sowing rate on yield using two way factorial treatment structure in an RCBD. Columns are treatment number, stock, rate, block, and yield.
• ratwt.txt Weight gain in rats fed diets with different protein levels and types. Design is a CRD. Columns are treatment #, level, type, and weight gain (in gm).
• range.txt Fertilizer effects on a pasture grass. The response variable is the PO4 concentration in leaf tissue. Variables are the treatment name, block #, PO4 conc., and a treatment code (a - e). The treatment codes identify the treatment groups, but the order is a bit more natural than the treatment name variable.
• range2.txt Same data as range.txt, but labeled for use in a 2 way ANOVA. Variables are Nrate (3 levels: 0, 50 and 100) and P (- or +), block # and PO4 measurement.
• hyp.txt "Botched" study for HW 8. Study was designed as an RCBD with 10 blocks and 2 treatments, but many values are missing. Three columns: block, treatment (a or p) and response.
• alpine1.txt Seedling growth in alpine tundra. Columns are the experiment (only Penn in this data set), the species names (nive or grac), the site (Dry, Med, Wet), and the log transformed weight.
• Alpine2.txt Potentilla growth at two sites. Columns are place, species, site and log transformed weight.
• alfalfacut.txt Alfalfa cutting experiment. 3 varieties, 4 dates of last cutting in fall. Used as example for a split plot study.
• ryegrass.txt Effect of manuring and ryegrass strain on pasture productivity.
• splitcornu.txt Effect of water, P, and N on corn water use efficiency.
• asparagus.txt Asparagus cutting study. Four harvesting intensities for asparagus. Plots repeatedly measured in 3 years. Used as example of repeated measures analysis.
• topsin.txt Topsin data set for midterm II. Columns are the shelf, a treatment number (1 through 15), the name of the fungal isolate, the concentration of fungicide, and the growth.
• bison.txt Bison data set for midterm II. Columns are the geographic area for a site (N, M, S), the age of the site (E, T, L), the site number (1 through 22), a code for the specific bone (a, b, c, or d), and log transformed length of the bone.
• serum.txt Data on serum glucose concentration over time.
• trigly.txt Data on triglyceride measurements, used as example of repeated studies.
• constipat.txt Data on
• camp.txt Data on cAMP, used as class example of ANCOVA for baseline data.
• heart.txt Data on effects of 2 drugs and a placebo on heart rate.
• wheatrep.txt Data from multi-location trial of wheat.