Show me the errors you didn’t look for!

# Show me the errors you didn’t look for!
### Claus Thorn Ekstrøm<br>UCPH Biostatistics<br>.small[<a href="mailto:ekstrom@sund.ku.dk">ekstrom@sund.ku.dk</a>]
### useR, July 6th 2017<br>.small[Slides @ <a href="www.biostatistics.dk/talks/">biostatistics.dk/talks/</a>]

---

background-image: url(pics/manbeer.jpg)
background-size: 100%
class: middle, center

# The RESCueH project

---

# Timeline followback (TLFB)

```
  day1 day2 day3
1   14   NA   NA
2   10   NA   99
3   19   14   40
4    5    7    7
```

---

# Timeline followback (TLFB)

```
  day1 day2 day3
1   14   NA   NA
2   10   NA   99
3   19   14   40
4    5    7    7
5   12    5    1
6   11   88    4
```

---

# Monthly Alcohol units

---

# Reproducible research

What **didn't** we check?

- Need **experts in relevant field**
- Merge existing databases into megadatabases
- New technologies revive old data

---

```
   colour int string    numeric uniq      ident
3     red   1      a -0.8356286    3 Irrelevant
4     red   2      a  1.5952808    4 Irrelevant
5     red   2      a  0.3295078    5 Irrelevant
6     red   6      b -0.8204684    6 Irrelevant
7     red   6      b  0.4874291    7 Irrelevant
8     red   6      b  0.7383247    8 Irrelevant
9     red 999      c  0.5757814    9 Irrelevant
10    red  NA      c -0.3053884   10 Irrelevant
11   blue   4      c  1.5117812   11 Irrelevant
12   blue  82      .  0.3898432   12 Irrelevant
13   blue  NA        -0.6212406   13 Irrelevant
14   <NA> NaN  other -2.2146999   14 Irrelevant
15   <NA>   5  OTHER  1.1249309   15 Irrelevant
```
]

---

# dataMaid

```r
library(dataMaid)
data(toyData)
*clean(toyData)
```

Documentation to be **read** and **evaluated** by a human.

See [github.com/ekstroem/dataMaid](github.com/ekstroem/dataMaid) for more info. Stable version on CRAN.

---

# Flowchart

---

# Part 1: Data cleaning summary

---

# Part 2: Summary table

---

# Part 3: Variable list

---

---

# Using dataMaid interactively

```r
> check(toyData$int)  # Individual check
$identifyMissing
The following suspected missing value codes 
    enter as regular values: 999, NaN.
$identifyOutliers
Note that the following possible outlier values 
    were detected: 82, 999.

> check(toyData$int, 
        numericChecks = "identifyMissing")
```
]

---

# Using dataMaid interactively

```r
visualize(toyData$int)
```
]
]

---

# Using dataMaid interactively

```r
summarize(toyData$int)
```

```
     Feature                   Result    
[1,] "Variable type"           "numeric" 
[2,] "Number of missing obs."  "3 (20 %)"
[3,] "Number of unique values" "8"       
[4,] "Median"                  "4.5"     
[5,] "1st and 3rd quartiles"   "1.75; 6" 
[6,] "Min. and max."           "1; 999"  
```
]

---

# Using dataMaid interactively

```r
> allSummaryFunctions()

-------------------------------------------------------------
name         description          classes                    
------------ -------------------- ---------------------------
centralValue Compute median       character, Date, factor,   
             or mode              integer, labelled, logical,
                                  numeric

countMissing Compute ratio of     character, Date, factor,   
             missing obs.         integer, labelled, logical,
                                  numeric

minMax       Find min and max     integer, numeric, Date     
             values

quartiles    Compute 1st and 3rd quartiles Date, integer, numeric

uniqueValues Count number of unique values character, Date, factor,   
                                           integer, labelled, logical,
                                           numeric

variableType Data class of variable        character, Date, factor,   
                                           integer, labelled, logical,
                                           numeric                    
----------------------------------------------------------------------
```
]

---

# Extending dataMaid

```r
isSSN <- function(v, nMax = NULL, ...) {
  out <- list(problem = FALSE, message = "")
  if (class(v) %in% c("character", "factor", 
                      "labelled")) {
    if (any(grep("\\d{3}-\\d{2}-\\d{4}", v))) {
      out$problem <- TRUE
      out$message <- 
        "Warning: Seems to contain SSNs."
    }  }
  out }
```
]
---

# Adding the function

```r
DF <- data.frame(ids=c("111-22-3333","123-45-6789", 
                       "111-22-3333"),
                 id2=c("111223333", "123456789", 
                       "4728491283"), 
                 stringsAsFactors=FALSE)

clean(DF,
  characterChecks = c("isSSN"))
```

```
Warning: Seems to contain SSNs.
```
]

---

# In summary

# 42

*   We **need** reproducible research to ensure that we can document exactly what we have done --- and also what we **haven't** done.
*   `dataMaid` addresses part of that requirement.
*   **You** are your worst collaborator.