Script won't run (Error in terms.formula(formula, data = data) : 'data' argument is of the wrong type

Question

I have run the script below numerous times and it has worked until this morning, when it suddenly produced the error message:

(Error in terms.formula(formula, data = data) : 'data' argument is of the wrong type.

I have not changed anything and I need to find out why it suddenly doesn't seem to work. Previous answers to similar questions have not helped.

My data:

DPUT(harvest2)
structure(list(Year = c(1971, 1972, 1973, 1974, 1975, 1976, 1977, 
1978, 1979, 1980, 1981, 1982, 1983, 1984, 1985, 1986, 1987, 1988, 
1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 
2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 
2011, 2012, 2013, 2014, 2015, 2016), Count = c(750, 757, 592, 
693, 667, 757, 719, 670, 733, 796, 923, 921, 944, 1010, 825, 
762, 825, 844, 809, 830, 768, 823, 749, 675, 700, 637, 708, 697, 
754, 694, 636, 717, 786, 731, 769, 732, 710, 610.5, 593, 529, 
664, 788, 731, 644, 653, 656), SexRat = c(1.91812865497076, 2.34567901234568, 
1.69178082191781, 1.46766169154229, 1.30396475770925,     
1.4364406779661, 1.32098765432099, 1.48584905660377, 1.5906976744186, 
1.91414141414141, 1.48905109489051, 1.61382113821138, 1.52380952380952, 
1.87777777777778, 1.75438596491228, 1.6695652173913, 1.81566820276498, 
1.79295154185022, 1.85024154589372, 1.75446428571429, 1.83163265306122, 
1.92857142857143, 1.76635514018692, 1.5, 2.26190476190476,     1.76704545454545,
2.38125, 1.80924855491329, 2.33333333333333, 1.81182795698925, 
2.20446096654275, 2.02790697674419, 2.1140350877193, 2.05, 2.20183486238532, 
1.90983606557377, 2.02262443438914, 1.75116279069767, 1.86842105263158, 
1.87951807228916, 2.08542713567839, 2.01724137931034, 1.95833333333333, 
1.81165919282511, 2.12135922330097, 1.97260273972603)), class = "data.frame", 
row.names = c(NA, -46L))

My script:

# Function for the equation

lm_eqn = function(df){
  m = lm(y ~ poly(x, 3), df) #3rd degree polynomial
  eq <- substitute(italic(y) == a + b %.% italic(x)*","~~italic(r)^2~"="~r2,
                   list(a = format(coef(m)[1], digits = 2),
                        b = format(coef(m)[2], digits = 2),
                        r2 = format(summary(m)$r.squared, digits = 4)))

  as.character(as.expression(eq))
}

# Make the plot

library(ggplot2)
ggplot(harvest2, aes(x = Year, y = Count)) +  
  scale_y_continuous(minor_breaks = seq(500, 1100, by = 50), 
                     breaks = seq(500, 1100, by = 100),
                     limits = c(500, 1100), expand = c(0, 0)) +  
  scale_x_continuous(minor_breaks = seq(1970, 2018, by = 1), 
                     breaks = seq(1970, 2018, by = 5), limits = c(1970, 2018)) +
  geom_point(stat = 'identity', size=2) +
  stat_smooth(method = "lm", se = TRUE, fill = NA, size = 1.3,
              formula = y ~ poly(x, 3, raw = TRUE), col = "red") +
  annotate("text", x = 1975, y = 1075, label = lm_eqn(df), 
           hjust = 0, size = 3.5, parse = TRUE) +
  xlab(" ") + 
  ylab("Count") +
  theme_light() +
  ggtitle(" ")

Any help much appreciated.

Answer 1

How about using stat_poly_eq from the ggpmisc package? See this if you want to separate the equation and R2 into two lines.

library(ggplot2)
library(ggpmisc)

# define formula
formula1 <- y ~ poly(x, 3, raw = TRUE)

ggplot(harvest2, aes(x = Year, y = Count)) +
  scale_y_continuous(
    minor_breaks = seq(500, 1100, by = 50), breaks = seq(500, 1100, by = 100),
    limits = c(500, 1100), expand = c(0, 0)) +
  scale_x_continuous(
    minor_breaks = seq(1970, 2018, by = 1), breaks = seq(1970, 2018, by = 5),
    limits = c(1970, 2018)) +
  geom_point(stat = "identity", size = 2) +
  stat_smooth(
    method = "lm", se = TRUE, fill = NA, size = 1.3,
    formula = formula1, col = "red") +
  # show the equation and R2
  stat_poly_eq(aes(label = paste(..eq.label.., ..rr.label.., sep = "~~~~")),
    label.x.npc = "left", label.y.npc = "top",
    formula = formula1, parse = TRUE, size = 5) +
  xlab(" ") + ylab("Count") +
  theme_light() +
  ggtitle(" ")

^{Created on 2019-02-12 by the reprex package (v0.2.1.9000)}

Answer 2

The answer from @Tung solves the problem, and Tung's comment hints at what is wrong in the original code. I expand on this below.

Most likely the problem is in the use of df in the statement below:

annotate("text", x = 1975, y = 1075, label = lm_eqn(df), hjust = 0, size = 3.5, parse = TRUE)

df is the name of the parameter not the argument needed, this should be for the equation to match the plotted curve the same data as passed to ggplot :

annotate("text", x = 1975, y = 1075, label = lm_eqn(df = harvest2), hjust = 0, size = 3.5, parse = TRUE)

The error and possible wrong equation coefficients/R^2 value will vary depending on whether or not there is a variable named df in the search path and what object is stored in it.

@Tung's code ensures that there is no mismatch between equation and plotted curve by defining the model formula only once and storing it in a variable as well as passing the data as argument only once.