Load the libraries and dataset

library(dplyr)
library(ggplot2)
library(ggfortify)
growth.moo <- read.csv('/Users/gerrisummers/Documents/Biostats4259a/Data/growth.csv')
glimpse(growth.moo)
## Rows: 48
## Columns: 3
## $ supplement <chr> "supergain", "supergain", "supergain", "supergain", "contro…
## $ diet       <chr> "wheat", "wheat", "wheat", "wheat", "wheat", "wheat", "whea…
## $ gain       <dbl> 17.37125, 16.81489, 18.08184, 15.78175, 17.70656, 18.22717,…

Loads the libraries and the data, and glimpse() shows its structure.

growth.moo <- mutate(growth.moo,
                     diet = factor(diet),
                     supplement = factor(supplement))

Use levels function to see the levels associated with a factor

levels(growth.moo$diet)
## [1] "barley" "oats"   "wheat"
levels(growth.moo$supplement)
## [1] "agrimore"  "control"   "supergain" "supersupp"

Relevel the order from control

growth.moo <-
  mutate(growth.moo,
       supplement = relevel(supplement, ref = "control"))

levels(growth.moo$supplement)
## [1] "control"   "agrimore"  "supergain" "supersupp"
levels(growth.moo$diet)
## [1] "barley" "oats"   "wheat"

Determine if the design is balanced

growth.moo %>%
  group_by(diet, supplement) %>%
  summarise(n = n())
## # A tibble: 12 × 3
## # Groups:   diet [3]
##    diet   supplement     n
##    <fct>  <fct>      <int>
##  1 barley control        4
##  2 barley agrimore       4
##  3 barley supergain      4
##  4 barley supersupp      4
##  5 oats   control        4
##  6 oats   agrimore       4
##  7 oats   supergain      4
##  8 oats   supersupp      4
##  9 wheat  control        4
## 10 wheat  agrimore       4
## 11 wheat  supergain      4
## 12 wheat  supersupp      4

Counts cows in each diet × supplement group, and it is balanced.

Finding outliers (Zuur)

growth.moo %>%
  mutate(order = row_number()) %>%
  ggplot(aes(x = gain, y = order)) +
  geom_point() +
  theme_bw()

Using a Cleveland dotplot there is no evidence of outliers in weight gain

Homogeneity of variance (Zuur)

ggplot(growth.moo, aes(x = supplement, y = gain, fill = diet)) +
  geom_boxplot() +
  theme_bw()

The spread of weight gain seems similar across groups, but barley groups appear slightly more variable to the others

Calculate the mean and sd for all 12 combinations

sumMoo <- growth.moo %>%
  group_by(diet, supplement) %>%
    summarise(
      meanGrow = mean(gain),
      seGrow = sd(gain)/sqrt(n())
    )
sumMoo
## # A tibble: 12 × 4
## # Groups:   diet [3]
##    diet   supplement meanGrow seGrow
##    <fct>  <fct>         <dbl>  <dbl>
##  1 barley control        23.3  0.703
##  2 barley agrimore       26.3  0.919
##  3 barley supergain      22.5  0.771
##  4 barley supersupp      25.6  1.06 
##  5 oats   control        20.5  0.506
##  6 oats   agrimore       23.3  0.613
##  7 oats   supergain      19.7  0.349
##  8 oats   supersupp      21.9  0.413
##  9 wheat  control        17.4  0.460
## 10 wheat  agrimore       19.6  0.710
## 11 wheat  supergain      17.0  0.485
## 12 wheat  supersupp      19.7  0.475

ggplot() interaction plot of sumMoo

ggplot(sumMoo, aes(x = supplement, y = meanGrow, colour = diet, group = diet)) +
  geom_point() +
  geom_line() +
  geom_errorbar(aes(ymin = meanGrow - seGrow,
                     ymax = meanGrow + seGrow), width = 0.1) +
  theme_bw()

Barley gives the highest weight gain and wheat the lowest under every supplement, and the lines are roughly parallel, which suggests there is no interaction between diet and supplement.

Make a linear model using lm() function. The “~” defines the relationship between variables and lm() makes the linear model and autoplot() draws a plot for an objective

model_cow <- lm(gain ~ diet * supplement, data = growth.moo)

autoplot(model_cow, smooth.colour = NA)

The residuals are evenly scattered around zero, the points fall mostly close to the line in the Q-Q plot with slight curving at the tails, and there is no clear fan shape in the Scale-Location plot (barley groups at high fitted values are slightly more spread, as in the boxplot), so the assumptions of the model appear reasonably met.

Using anova() to produce and ANOVA table listing sums of squares, mean squares, F-values and p-values. Then using summary() to return a table of coefficients

anova(model_cow)
## Analysis of Variance Table
## 
## Response: gain
##                 Df  Sum Sq Mean Sq F value    Pr(>F)    
## diet             2 287.171 143.586 83.5201 2.999e-14 ***
## supplement       3  91.881  30.627 17.8150 2.952e-07 ***
## diet:supplement  6   3.406   0.568  0.3302    0.9166    
## Residuals       36  61.890   1.719                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

After reviewing the ANOVA table, the interaction between diet and supplement on the weight gain of the cows is not significant (F=0.3302, df=6,36, p=0.92). We fail to reject the null hypothesis, as there is no evidence of an interaction between diet and supplement; the effect of supplement on weight gain does not depend on diet. Diet has a significant effect on weight gain (F = 83.52, df = 2, 36, p < 0.001), and supplement has a significant effect on weight gain (F = 17.82, df = 3, 36, p < 0.001).

summary(model_cow)
## 
## Call:
## lm(formula = gain ~ diet * supplement, data = growth.moo)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -2.48756 -1.00368 -0.07452  1.03496  2.68069 
## 
## Coefficients:
##                                 Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                   23.2966499  0.6555863  35.536  < 2e-16 ***
## dietoats                      -2.8029851  0.9271390  -3.023  0.00459 ** 
## dietwheat                     -5.8911317  0.9271390  -6.354 2.34e-07 ***
## supplementagrimore             3.0518277  0.9271390   3.292  0.00224 ** 
## supplementsupergain           -0.8305263  0.9271390  -0.896  0.37631    
## supplementsupersupp            2.2786527  0.9271390   2.458  0.01893 *  
## dietoats:supplementagrimore   -0.2471088  1.3111726  -0.188  0.85157    
## dietwheat:supplementagrimore  -0.8182729  1.3111726  -0.624  0.53651    
## dietoats:supplementsupergain  -0.0001351  1.3111726   0.000  0.99992    
## dietwheat:supplementsupergain  0.4374395  1.3111726   0.334  0.74060    
## dietoats:supplementsupersupp  -0.9120830  1.3111726  -0.696  0.49113    
## dietwheat:supplementsupersupp -0.0158299  1.3111726  -0.012  0.99043    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.311 on 36 degrees of freedom
## Multiple R-squared:  0.8607, Adjusted R-squared:  0.8182 
## F-statistic: 20.22 on 11 and 36 DF,  p-value: 3.295e-12

The intercept is the mean weight gain for cows on barley with the control supplement, and every other coefficient is the difference from that reference group.

Loading the data for ANCOVA

limp <- read.csv('/Users/gerrisummers/Documents/Biostats4259a/Data/limpet.csv')

limp <- mutate(limp,
               SEASON = factor(SEASON))

glimpse(limp)
## Rows: 24
## Columns: 3
## $ DENSITY <int> 8, 8, 8, 8, 8, 8, 15, 15, 15, 15, 15, 15, 30, 30, 30, 30, 30, …
## $ SEASON  <fct> spring, spring, spring, summer, summer, summer, spring, spring…
## $ EGGS    <dbl> 2.875, 2.625, 1.750, 2.125, 1.500, 1.875, 2.600, 1.866, 2.066,…

Checking if data is balanced (Zuur)

limp %>%
  group_by(SEASON) %>%
  summarise(n = n())
## # A tibble: 2 × 2
##   SEASON     n
##   <fct>  <int>
## 1 spring    12
## 2 summer    12

Counts match (12) and data is balanced

Looking for outliers (Zuur)

limp %>%
  mutate(order = row_number()) %>%
  ggplot(aes(x = EGGS, y = order)) +
  geom_point() +
  theme_bw()

limp %>%
  mutate(order = row_number()) %>%
  ggplot(aes(x = DENSITY, y = order)) +
  geom_point() +
  theme_bw()

No outliers detected in dataset.

Plot window

ggplot(limp, aes(x = DENSITY, y = EGGS, colour = SEASON)) + geom_point() +
  scale_colour_manual(values = c(spring="green",summer="red")) +
theme_bw()

Egg production decreases as limpet density increases, spring limpets produce more eggs, and the two seasons appear to have similar slopes.

Specifying a general linear model from the limp

limp.mod <- lm(EGGS ~ DENSITY * SEASON, data = limp)

names(limp.mod)
##  [1] "coefficients"  "residuals"     "effects"       "rank"         
##  [5] "fitted.values" "assign"        "qr"            "df.residual"  
##  [9] "contrasts"     "xlevels"       "call"          "terms"        
## [13] "model"

Create diagnostic graphs using autoplot() which draws a plot for an objective from limp.mod

autoplot(limp.mod, smooth.colour = NA)

The residuals are evenly scattered around zero, the points fall close to the line in the Q-Q plot, there is no fan shape in the Scale-Location plot, and no points have unusually high leverage, so the assumptions of the model appear to be met.

Use anova() and summary() to view interactions of density:season and eggs produced

anova(limp.mod)
## Analysis of Variance Table
## 
## Response: EGGS
##                Df Sum Sq Mean Sq F value    Pr(>F)    
## DENSITY         1 5.0241  5.0241 30.1971 2.226e-05 ***
## SEASON          1 3.2502  3.2502 19.5350 0.0002637 ***
## DENSITY:SEASON  1 0.0118  0.0118  0.0711 0.7925333    
## Residuals      20 3.3275  0.1664                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

There is no evidence of an interaction between density and season (F = 0.0711, df = 1, 20, p = 0.79). Therefore we fail to reject the null hypothesis; the effect of density on egg production in limpets does not depend on the season. Density has a significant effect on egg production (F = 30.20, df = 1, 20, p < 0.001), and season has a significant effect on egg production (F = 19.54, df = 1, 20, p < 0.001).

Create summary table of limp.mod using summary() function

summary(limp.mod)
## 
## Call:
## lm(formula = EGGS ~ DENSITY * SEASON, data = limp)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.65468 -0.25021 -0.03318  0.28335  0.57532 
## 
## Coefficients:
##                       Estimate Std. Error t value Pr(>|t|)    
## (Intercept)           2.664166   0.234118  11.380 3.45e-10 ***
## DENSITY              -0.033650   0.008259  -4.074 0.000591 ***
## SEASONsummer         -0.812282   0.331092  -2.453 0.023450 *  
## DENSITY:SEASONsummer  0.003114   0.011680   0.267 0.792533    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.4079 on 20 degrees of freedom
## Multiple R-squared:  0.7135, Adjusted R-squared:  0.6705 
## F-statistic:  16.6 on 3 and 20 DF,  p-value: 1.186e-05

The intercept and DENSITY coefficient are the intercept and slope for spring, SEASONsummer is how much the summer intercept differs from spring, and DENSITY:SEASONsummer is how much the summer slope differs from spring.

Build a factorial display of variables using the function expand.grid()

expand.grid(FRIST = c("A", "B"), SECOND = c(1,2))
##   FRIST SECOND
## 1     A      1
## 2     B      1
## 3     A      2
## 4     B      2

Use predict() to generate fitted values from a model

predict(limp.mod)
##         1         2         3         4         5         6         7         8 
## 2.3949692 2.3949692 2.3949692 1.6075953 1.6075953 1.6075953 2.1594217 2.1594217 
##         9        10        11        12        13        14        15        16 
## 2.1594217 1.3938428 1.3938428 1.3938428 1.6546769 1.6546769 1.6546769 0.9358016 
##        17        18        19        20        21        22        23        24 
## 0.9358016 0.9358016 1.1499321 1.1499321 1.1499321 0.4777604 0.4777604 0.4777604

Make new DENSITY values which we request predictions and checked it worked with head()

new.x <- expand.grid(DENSITY =
                       seq(from = 8, to =45, length.out = 10), SEASON = levels(limp$SEASON))
head(new.x)
##    DENSITY SEASON
## 1  8.00000 spring
## 2 12.11111 spring
## 3 16.22222 spring
## 4 20.33333 spring
## 5 24.44444 spring
## 6 28.55556 spring

Generate fits and confidence interval at new .x values

new.y <- predict(limp.mod, newdata=new.x,
                 interval = 'confidence')
head(new.y)
##        fit      lwr      upr
## 1 2.394969 2.019285 2.770654
## 2 2.256632 1.931230 2.582034
## 3 2.118294 1.834274 2.402315
## 4 1.979957 1.724062 2.235852
## 5 1.841619 1.595998 2.087241
## 6 1.703282 1.447918 1.958646

Rename fit to predict EGGS using rename()

addThese <- data.frame(new.x, new.y)
addThese <- rename(addThese, EGGS= fit)

head(addThese)
##    DENSITY SEASON     EGGS      lwr      upr
## 1  8.00000 spring 2.394969 2.019285 2.770654
## 2 12.11111 spring 2.256632 1.931230 2.582034
## 3 16.22222 spring 2.118294 1.834274 2.402315
## 4 20.33333 spring 1.979957 1.724062 2.235852
## 5 24.44444 spring 1.841619 1.595998 2.087241
## 6 28.55556 spring 1.703282 1.447918 1.958646

Getting the predicted number of eggs for each season at the mean limpet density

# new.w using density to establish a mean
new.x <- expand.grid(DENSITY=mean(limp$DENSITY),
                     SEASON = levels(limp$SEASON))

# predicting 
predEgg <- predict(limp.mod, newdata = new.x)

# Housekeeping
EggAtMeanDens <- data.frame(new.x, predEgg)
head(EggAtMeanDens)
##   DENSITY SEASON predEgg
## 1    24.5 spring 1.83975
## 2    24.5 summer 1.10375

At the mean limpet density (24.5), the model predicts 1.84 eggs in spring and 1.10 eggs in summer, so spring limpets produce more eggs than summer limpets at the same density.

Using ggplot() to plot the model

ggplot(limp, aes(x = DENSITY, y = EGGS, colour = SEASON)) + geom_point(size =5) + geom_smooth(data = addThese, aes(ymin = lwr, ymax = upr, fill = SEASON), stat = 'identity') + scale_colour_manual(values = c(spring="green",summer="red")) +
  scale_fill_manual(values = c(spring="green",summer="red"))+
  theme_bw()

The fitted lines show egg production decreasing with density in both seasons, with parallel slopes (no interaction) and spring consistently higher than summer.