0

I've used the exact same code for several different spreadsheets, but I seem to be getting an error for this specific spreadsheet and I cannot figure out why. This is the code attached below:

> library(readxl)
> library(sjPlot)
> data <- read_excel("C:\\Users\\hadik\\Downloads\\sentimprovdepression.xlsx")
> head(data)
# A tibble: 6 × 16
  `Patient Start Positi…` `Patient Start…` `Patient Start…` Type  `Therapist Sta…` `Therapist Sta…`
                    <dbl>            <dbl>            <dbl> <chr>            <dbl>            <dbl>
1                    0.33             0.32             0.45 Pres…             0.35             0.31
2                    0.34             0.32             0.43 Pres…             0.33             0.32
3                    0.33             0.33             0.43 Pres…             0.34             0.32
4                    0.32             0.32             0.44 Pres…             0.33             0.31
5                    0.35             0.32             0.43 Pres…             0.33             0.32
6                    0.35             0.32             0.42 Pres…             0.34             0.31
# … with 10 more variables: `Therapist Start Neutral Sentiment` <dbl>, Gender <chr>,
#   Ethnicity <chr>, Employment_Status <chr>, StartPHQ9 <dbl>, StartGAD7 <dbl>, StartWSAS <dbl>,
#   Treatment_sessions <dbl>, Age_group <chr>, Improvement <dbl>
> str(data)
tibble [12,805 × 16] (S3: tbl_df/tbl/data.frame)
 $ Patient Start Positive Sentiment  : num [1:12805] 0.33 0.34 0.33 0.32 0.35 0.35 0.35 0.33 0.35 0.34 ...
 $ Patient Start Negative Sentiment  : num [1:12805] 0.32 0.32 0.33 0.32 0.32 0.32 0.34 0.31 0.32 0.33 ...
 $ Patient Start Neutral Sentiment   : num [1:12805] 0.45 0.43 0.43 0.44 0.43 0.42 0.41 0.44 0.42 0.43 ...
 $ Type                              : chr [1:12805] "Prescribed Taking" "Prescribed Taking" "Prescribed Taking" "Prescribed Taking" ...
 $ Therapist Start Positive Sentiment: num [1:12805] 0.35 0.33 0.34 0.33 0.33 0.34 0.34 0.34 0.34 0.33 ...
 $ Therapist Start Negative Sentiment: num [1:12805] 0.31 0.32 0.32 0.31 0.32 0.31 0.32 0.32 0.32 0.31 ...
 $ Therapist Start Neutral Sentiment : num [1:12805] 0.43 0.45 0.44 0.45 0.45 0.45 0.43 0.44 0.44 0.45 ...
 $ Gender                            : chr [1:12805] "M" "M" "F" "F" ...
 $ Ethnicity                         : chr [1:12805] "White" "White" "White" "White" ...
 $ Employment_Status                 : chr [1:12805] "Employed" "Employed" "Other" "Other" ...
 $ StartPHQ9                         : num [1:12805] 21 5 20 23 21 18 22 12 21 11 ...
 $ StartGAD7                         : num [1:12805] 18 3 16 15 18 11 21 13 16 4 ...
 $ StartWSAS                         : num [1:12805] 20 8 30 37 32 23 36 25 36 14 ...
 $ Treatment_sessions                : num [1:12805] 7 5 15 9 13 9 10 13 10 13 ...
 $ Age_group                         : chr [1:12805] "36-45" "46-55" "18-25" "36-45" ...
 $ Improvement                       : num [1:12805] 1 0 0 0 1 1 0 1 1 0 ...
> ##Make sure to scale all the continuous data first##
> data$Gender <- as.factor(data$Gender)
> data$Ethnicity <- as.factor(data$Ethnicity)
> data$'Employment_Status' <- as.factor(data$'Employment_Status')
> data$'Improvement' <- as.factor(data$'Improvement')
> data$'Age_group' <- as.factor(data$'Age_group')
> data$'Type' <- as.factor(data$'Type')
> data$'StartPHQ9' <- as.integer(data$'StartPHQ9')
> data$'StartGAD7' <- as.integer(data$'StartGAD7')
> data$'StartWSAS' <- as.integer(data$'StartWSAS')
> data$'Treatment_sessions' <- as.integer(data$'Treatment_sessions')
> data$Gender <- relevel(data$Gender, ref = "F")
> data$Improvement <- as.factor(data$Improvement)
> str(data)
tibble [12,805 × 16] (S3: tbl_df/tbl/data.frame)
 $ Patient Start Positive Sentiment  : num [1:12805] 0.33 0.34 0.33 0.32 0.35 0.35 0.35 0.33 0.35 0.34 ...
 $ Patient Start Negative Sentiment  : num [1:12805] 0.32 0.32 0.33 0.32 0.32 0.32 0.34 0.31 0.32 0.33 ...
 $ Patient Start Neutral Sentiment   : num [1:12805] 0.45 0.43 0.43 0.44 0.43 0.42 0.41 0.44 0.42 0.43 ...
 $ Type                              : Factor w/ 3 levels "Not Prescribed",..: 3 3 3 3 3 3 3 3 3 3 ...
 $ Therapist Start Positive Sentiment: num [1:12805] 0.35 0.33 0.34 0.33 0.33 0.34 0.34 0.34 0.34 0.33 ...
 $ Therapist Start Negative Sentiment: num [1:12805] 0.31 0.32 0.32 0.31 0.32 0.31 0.32 0.32 0.32 0.31 ...
 $ Therapist Start Neutral Sentiment : num [1:12805] 0.43 0.45 0.44 0.45 0.45 0.45 0.43 0.44 0.44 0.45 ...
 $ Gender                            : Factor w/ 3 levels "F","M","Z": 2 2 1 1 2 2 1 1 1 1 ...
 $ Ethnicity                         : Factor w/ 3 levels "other","U/Z",..: 3 3 3 3 3 3 3 3 1 2 ...
 $ Employment_Status                 : Factor w/ 4 levels "Employed","Other",..: 1 1 2 2 1 1 1 1 1 1 ...
 $ StartPHQ9                         : int [1:12805] 21 5 20 23 21 18 22 12 21 11 ...
 $ StartGAD7                         : int [1:12805] 18 3 16 15 18 11 21 13 16 4 ...
 $ StartWSAS                         : int [1:12805] 20 8 30 37 32 23 36 25 36 14 ...
 $ Treatment_sessions                : int [1:12805] 7 5 15 9 13 9 10 13 10 13 ...
 $ Age_group                         : Factor w/ 6 levels "18-25","26-35",..: 3 4 1 3 3 3 2 1 3 2 ...
 $ Improvement                       : Factor w/ 2 levels "0","1": 2 1 1 1 2 2 1 2 2 1 ...
> logistic <- glm(Improvement ~., data=data, family = "binomial")
> summary(logistic)

Call:
glm(formula = Improvement ~ ., family = "binomial", data = data)

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-2.7150  -1.0930   0.6314   0.9430   1.9979  

Coefficients:
                                       Estimate Std. Error z value Pr(>|z|)    
(Intercept)                            2.760454   5.572445   0.495 0.620335    
`Patient Start Positive Sentiment`     7.340544   3.513784   2.089 0.036701 *  
`Patient Start Negative Sentiment`   -15.468152   3.730949  -4.146 3.38e-05 ***
`Patient Start Neutral Sentiment`      4.414449   4.052710   1.089 0.276040    
TypePrescribed Not Taking             -0.260718   0.087927  -2.965 0.003025 ** 
TypePrescribed Taking                 -0.173552   0.042759  -4.059 4.93e-05 ***
`Therapist Start Positive Sentiment`   0.355103   3.750067   0.095 0.924559    
`Therapist Start Negative Sentiment`  -7.829533   4.358366  -1.796 0.072425 .  
`Therapist Start Neutral Sentiment`   -2.962382   4.514409  -0.656 0.511692    
GenderM                               -0.058465   0.043917  -1.331 0.183101    
GenderZ                               -0.278082   0.302149  -0.920 0.357392    
EthnicityU/Z                           0.297006   0.083974   3.537 0.000405 ***
EthnicityWhite                         0.045357   0.076317   0.594 0.552294    
Employment_StatusOther                -0.348494   0.050218  -6.940 3.93e-12 ***
Employment_StatusU/Z                  -0.345905   0.099468  -3.478 0.000506 ***
Employment_StatusUnemployed           -0.159407   0.081467  -1.957 0.050380 .  
StartPHQ9                              0.049765   0.005029   9.896  < 2e-16 ***
StartGAD7                              0.089085   0.005057  17.618  < 2e-16 ***
StartWSAS                             -0.051228   0.002966 -17.272  < 2e-16 ***
Treatment_sessions                     0.171704   0.006444  26.646  < 2e-16 ***
Age_group26-35                         0.175195   0.052575   3.332 0.000861 ***
Age_group36-45                         0.260732   0.058407   4.464 8.04e-06 ***
Age_group46-55                         0.391842   0.066858   5.861 4.61e-09 ***
Age_group56-65                         0.233992   0.089346   2.619 0.008820 ** 
Age_group66+                           0.669370   0.157014   4.263 2.02e-05 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 17071  on 12804  degrees of freedom
Residual deviance: 15256  on 12780  degrees of freedom
AIC: 15306

Number of Fisher Scoring iterations: 3

> df <- get_model_data(logistic, type = NULL)
> df
                                 term     estimate   std.error conf.level     conf.low
1    Patient Start Positive Sentiment 1.541550e+03 3.513784263       0.95 1.574052e+00
2    Patient Start Negative Sentiment 1.915432e-07 3.730949432       0.95 1.277841e-10
3     Patient Start Neutral Sentiment 8.263631e+01 4.052710244       0.95 2.934254e-02
4           TypePrescribed Not Taking 7.704983e-01 0.087927189       0.95 6.485268e-01
5               TypePrescribed Taking 8.406734e-01 0.042758555       0.95 7.730920e-01
6  Therapist Start Positive Sentiment 1.426328e+00 3.750066645       0.95 9.165515e-04
7  Therapist Start Negative Sentiment 3.978113e-04 4.358366114       0.95 7.759419e-08
8   Therapist Start Neutral Sentiment 5.169562e-02 4.514409264       0.95 7.426456e-06
9                             GenderM 9.432113e-01 0.043916768       0.95 8.654201e-01
10                            GenderZ 7.572346e-01 0.302149420       0.95 4.188325e-01
11                       EthnicityU/Z 1.345824e+00 0.083973921       0.95 1.141588e+00
12                     EthnicityWhite 1.046401e+00 0.076316816       0.95 9.010260e-01
13             Employment_StatusOther 7.057500e-01 0.050217905       0.95 6.395954e-01
14               Employment_StatusU/Z 7.075795e-01 0.099467542       0.95 5.822485e-01
15        Employment_StatusUnemployed 8.526489e-01 0.081466610       0.95 7.268181e-01
16                          StartPHQ9 1.051024e+00 0.005028766       0.95 1.040716e+00
17                          StartGAD7 1.093174e+00 0.005056586       0.95 1.082393e+00
18                          StartWSAS 9.500620e-01 0.002966029       0.95 9.445550e-01
19                 Treatment_sessions 1.187326e+00 0.006443815       0.95 1.172425e+00
20                     Age_group26-35 1.191479e+00 0.052575083       0.95 1.074816e+00
21                     Age_group36-45 1.297880e+00 0.058407189       0.95 1.157493e+00
22                     Age_group46-55 1.479703e+00 0.066857937       0.95 1.297971e+00
23                     Age_group56-65 1.263634e+00 0.089345836       0.95 1.060645e+00
24                       Age_group66+ 1.953006e+00 0.157013623       0.95 1.435666e+00
      conf.high    statistic df.error       p.value p.stars   p.label group xpos   xmin   xmax
1  1.509720e+06   2.08907076      Inf  3.670135e-02       * 1541.55 *   pos   24 23.825 24.175
2  2.871155e-04  -4.14590245      Inf  3.384779e-05     ***  0.00 ***   neg   23 22.825 23.175
3  2.327256e+05   1.08925854      Inf  2.760399e-01            82.64    pos   22 21.825 22.175
4  9.154096e-01  -2.96515556      Inf  3.025299e-03      **   0.77 **   neg   21 20.825 21.175
5  9.141627e-01  -4.05888329      Inf  4.930795e-05     ***  0.84 ***   neg   20 19.825 20.175
6  2.219637e+03   0.09469253      Inf  9.245590e-01             1.43    pos   19 18.825 19.175
7  2.039507e+00  -1.79643758      Inf  7.242495e-02             0.00    neg   18 17.825 18.175
8  3.598536e+02  -0.65620596      Inf  5.116916e-01             0.05    neg   17 16.825 17.175
9  1.027995e+00  -1.33126781      Inf  1.831009e-01             0.94    neg   16 15.825 16.175
10 1.369054e+00  -0.92034668      Inf  3.573916e-01             0.76    neg   15 14.825 15.175
11 1.586598e+00   3.53688776      Inf  4.048717e-04     ***  1.35 ***   pos   14 13.825 14.175
12 1.215232e+00   0.59432601      Inf  5.522941e-01             1.05    pos   13 12.825 13.175
13 7.787471e-01  -6.93964197      Inf  3.930949e-12     ***  0.71 ***   neg   12 11.825 12.175
14 8.598886e-01  -3.47756880      Inf  5.059832e-04     ***  0.71 ***   neg   11 10.825 11.175
15 1.000264e+00  -1.95672157      Inf  5.038021e-02             0.85    neg   10  9.825 10.175
16 1.061435e+00   9.89613991      Inf  4.326530e-23     ***  1.05 ***   pos    9  8.825  9.175
17 1.104062e+00  17.61765088      Inf  1.803395e-69     ***  1.09 ***   pos    8  7.825  8.175
18 9.556011e-01 -17.27157987      Inf  7.700516e-67     ***  0.95 ***   neg    7  6.825  7.175
19 1.202417e+00  26.64628345      Inf 1.976286e-156     ***  1.19 ***   pos    6  5.825  6.175
20 1.320804e+00   3.33228522      Inf  8.613593e-04     ***  1.19 ***   pos    5  4.825  5.175
21 1.455295e+00   4.46404717      Inf  8.042584e-06     ***  1.30 ***   pos    4  3.825  4.175
22 1.686880e+00   5.86081093      Inf  4.606122e-09     ***  1.48 ***   pos    3  2.825  3.175
23 1.505471e+00   2.61894147      Inf  8.820308e-03      **   1.26 **   pos    2  1.825  2.175
24 2.656770e+00   4.26313269      Inf  2.015807e-05     ***  1.95 ***   pos    1  0.825  1.175
> exp(logistic$coefficients)
                         (Intercept)   `Patient Start Positive Sentiment` 
                        1.580702e+01                         1.541550e+03 
  `Patient Start Negative Sentiment`    `Patient Start Neutral Sentiment` 
                        1.915432e-07                         8.263631e+01 
           TypePrescribed Not Taking                TypePrescribed Taking 
                        7.704983e-01                         8.406734e-01 
`Therapist Start Positive Sentiment` `Therapist Start Negative Sentiment` 
                        1.426328e+00                         3.978113e-04 
 `Therapist Start Neutral Sentiment`                              GenderM 
                        5.169562e-02                         9.432113e-01 
                             GenderZ                         EthnicityU/Z 
                        7.572346e-01                         1.345824e+00 
                      EthnicityWhite               Employment_StatusOther 
                        1.046401e+00                         7.057500e-01 
                Employment_StatusU/Z          Employment_StatusUnemployed 
                        7.075795e-01                         8.526489e-01 
                           StartPHQ9                            StartGAD7 
                        1.051024e+00                         1.093174e+00 
                           StartWSAS                   Treatment_sessions 
                        9.500620e-01                         1.187326e+00 
                      Age_group26-35                       Age_group36-45 
                        1.191479e+00                         1.297880e+00 
                      Age_group46-55                       Age_group56-65 
                        1.479703e+00                         1.263634e+00 
                        Age_group66+ 
                        1.953006e+00 
> exp(confint(logistic))
Waiting for profiling to be done...
                                            2.5 %       97.5 %
(Intercept)                          2.864647e-04 8.809859e+05
`Patient Start Positive Sentiment`   1.567316e+00 1.507918e+06
`Patient Start Negative Sentiment`   1.266134e-10 2.851379e-04
`Patient Start Neutral Sentiment`    2.909615e-02 2.317040e+05
TypePrescribed Not Taking            6.487173e-01 9.157834e-01
TypePrescribed Taking                7.730574e-01 9.141332e-01
`Therapist Start Positive Sentiment` 9.140112e-04 2.217053e+03
`Therapist Start Negative Sentiment` 7.750747e-08 2.040810e+00
`Therapist Start Neutral Sentiment`  7.394264e-06 3.590291e+02
GenderM                              8.654807e-01 1.028082e+00
GenderZ                              4.198807e-01 1.379708e+00
EthnicityU/Z                         1.141286e+00 1.586287e+00
EthnicityWhite                       9.006407e-01 1.214808e+00
Employment_StatusOther               6.396016e-01 7.787699e-01
Employment_StatusU/Z                 5.823355e-01 8.601788e-01
Employment_StatusUnemployed          7.271343e-01 1.000799e+00
StartPHQ9                            1.040728e+00 1.061449e+00
StartGAD7                            1.082418e+00 1.104090e+00
StartWSAS                            9.445387e-01 9.555858e-01
Treatment_sessions                   1.172494e+00 1.202491e+00
Age_group26-35                       1.074829e+00 1.320841e+00
Age_group36-45                       1.157607e+00 1.455469e+00
Age_group46-55                       1.298312e+00 1.687380e+00
Age_group56-65                       1.061158e+00 1.506344e+00
Age_group66+                         1.439507e+00 2.665708e+00
> df <- df[-c(4,5,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24), ]#The numbers inside the parentheses indicates the number of rows you want removed
> df2 <- df
> label <- paste(df2$term)
> mean <- df2$estimate
> mean <- as.numeric(mean)
> lower <- df2$conf.low
> upper <- df2$conf.high
> sig <- df2$p.value
> stars <- df2$p.stars
> df <- data.frame(label,mean,lower,upper,sig,stars)
> df <- factor(df$label, levels=rev(df$label))
> df <- data.frame(label,mean,lower,upper,sig,stars)
> df$label <- factor(df$label, levels=rev(df$label))
> df
                               label         mean        lower        upper          sig stars
1   Patient Start Positive Sentiment 1.541550e+03 1.574052e+00 1.509720e+06 3.670135e-02     *
2   Patient Start Negative Sentiment 1.915432e-07 1.277841e-10 2.871155e-04 3.384779e-05   ***
3    Patient Start Neutral Sentiment 8.263631e+01 2.934254e-02 2.327256e+05 2.760399e-01      
4 Therapist Start Positive Sentiment 1.426328e+00 9.165515e-04 2.219637e+03 9.245590e-01      
5 Therapist Start Negative Sentiment 3.978113e-04 7.759419e-08 2.039507e+00 7.242495e-02      
6  Therapist Start Neutral Sentiment 5.169562e-02 7.426456e-06 3.598536e+02 5.116916e-01      
> df2
                                term     estimate std.error conf.level     conf.low    conf.high
1   Patient Start Positive Sentiment 1.541550e+03  3.513784       0.95 1.574052e+00 1.509720e+06
2   Patient Start Negative Sentiment 1.915432e-07  3.730949       0.95 1.277841e-10 2.871155e-04
3    Patient Start Neutral Sentiment 8.263631e+01  4.052710       0.95 2.934254e-02 2.327256e+05
6 Therapist Start Positive Sentiment 1.426328e+00  3.750067       0.95 9.165515e-04 2.219637e+03
7 Therapist Start Negative Sentiment 3.978113e-04  4.358366       0.95 7.759419e-08 2.039507e+00
8  Therapist Start Neutral Sentiment 5.169562e-02  4.514409       0.95 7.426456e-06 3.598536e+02
    statistic df.error      p.value p.stars   p.label group xpos   xmin   xmax
1  2.08907076      Inf 3.670135e-02       * 1541.55 *   pos   24 23.825 24.175
2 -4.14590245      Inf 3.384779e-05     ***  0.00 ***   neg   23 22.825 23.175
3  1.08925854      Inf 2.760399e-01            82.64    pos   22 21.825 22.175
6  0.09469253      Inf 9.245590e-01             1.43    pos   19 18.825 19.175
7 -1.79643758      Inf 7.242495e-02             0.00    neg   18 17.825 18.175
8 -0.65620596      Inf 5.116916e-01             0.05    neg   17 16.825 17.175
> library(dplyr)
> df3 <- arrange(df, -mean)
> df3$label <- reorder(df3$label, df3$mean)
> df3
                               label         mean        lower        upper          sig stars
1   Patient Start Positive Sentiment 1.541550e+03 1.574052e+00 1.509720e+06 3.670135e-02     *
2    Patient Start Neutral Sentiment 8.263631e+01 2.934254e-02 2.327256e+05 2.760399e-01      
3 Therapist Start Positive Sentiment 1.426328e+00 9.165515e-04 2.219637e+03 9.245590e-01      
4  Therapist Start Neutral Sentiment 5.169562e-02 7.426456e-06 3.598536e+02 5.116916e-01      
5 Therapist Start Negative Sentiment 3.978113e-04 7.759419e-08 2.039507e+00 7.242495e-02      
6   Patient Start Negative Sentiment 1.915432e-07 1.277841e-10 2.871155e-04 3.384779e-05   ***
> library(ggplot2)
> fp <- ggplot(df2, aes(label, mean, colour = mean>1)) +
+     
+     scale_colour_manual(name = 'mean>0', values = setNames(c('royalblue2','red3'),c(T, F))) +
+     
+     geom_pointrange(aes(ymin = lower, ymax = upper), size = 1) +
+     
+     geom_hline(yintercept=1, lty=2) +
+     
+     labs(title = "Interaction Between Medication Status and Sentiment for Improvement - Depression", x = "", y = "Odds Ratio") +
+     
+     coord_flip() +
+     
+     theme(legend.position="none") +
+     
+     ylim(0.0,10.0) +
+     
+     geom_text(aes(label=stars), position=position_dodge(width=0.9), hjust = -.35, vjust=.2, size = 6) +
+     
+     theme(axis.title = element_text(size = 14,colour = "black")) +
+     
+     theme(panel.background = element_rect(fill = "gray88")) +
+     
+     theme(plot.title = element_text(size = rel(1.5), hjust = 0.5)) +
+     
+     theme(axis.text=element_text(size=15,colour = "black"))
> 
> print(fp)
Warning messages:
1: Removed 2 rows containing missing values (geom_pointrange). 
2: Removed 2 rows containing missing values (geom_segment). 
3: Removed 2 rows containing missing values (geom_text). 
>

I've tried everything however the code just doesn't want to work. You can see the error in the last three lines and nothing I do seems to fix it.

halfer
  • 19,824
  • 17
  • 99
  • 186
Hadi
  • 13
  • 2
  • 1. Please modify your code so that you can provide us with a [minimal reproducible example](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) that creates your errors. We don't have access to your data files, so you'll want to use `dput` or otherwise give us a very simple toy dataset that produces the issue. Similarly, the example isn't *minimal* as it contains lots of extraneous code/printouts. (2) You're receiving warnings not errors (different things in R!). (3) You're overwriting the same variable over and over with `df <-` and it's not clear why.... – socialscientist Jul 26 '22 at 22:54
  • 2
    You have set `ylim` as `c(0, 10)` but not all the values are in this range. They are therefore removed by ggplot, which is what the warning is telling you. What happens when you remove ylim? – Allan Cameron Jul 26 '22 at 22:58
  • @AllanCameron removing it removed the warning messages however all the datapoints in the forestplot are zero... – Hadi Jul 26 '22 at 23:10
  • 1
    @Hadi are you sure they are all zero, or do they just _look_ zero because the upper confidence limit is 1.5 million for the first row? All the other numbers are several magnitudes lower, so will appear to be zero on a plot. – Allan Cameron Jul 26 '22 at 23:22
  • @AllanCameron that might be the case. Do you know if there is any way to pinpoint or perhaps annotate the datapoints with the exact x-value they're plotted on? I appreciate your help man. I really do. – Hadi Jul 27 '22 at 01:40

0 Answers0