search

nationalparkservice / DPchecker

DPchecker (Data Package checker) is a package with a series of functions for NPS data package authors and reviewers to check for internal consistency among data/meta data and with the data package standards
https://nationalparkservice.github.io/DPchecker/
Other
4 stars 1 forks source link

missing values are not behaving properly #137

Open RobLBaker opened 2 months ago

RobLBaker commented 2 months ago

getting reports that missing values such as "NULL" are causing issues if they are in numeric columns. This shouldn't be an issue. Haven't been able to get ahold of a dataset that contains this particular error yet, but it is worth testing further as multiple SCA data strike team members have experienced the issue.

RobLBaker commented 2 months ago

From the people who ran into this problem... maybe the issue is more about R not liking mixed data types than the DPchecker test:

"requested we use NULL for all missing values throughout the dataset, so that is what I did originally. I don't remember exactly what the wording of the warning was that came up from DPchecker, but it had the word "parse" or "parsing" in it. When I went on the github page that lists the functions in DPchecker and searched "parse" the function I landed on was Test data and metadata data formats match — test_dates_parse • DPchecker (nationalparkservice.github.io)". I think when I originally read the function name it I misread "dates" as "data" but when I called problems() for this function it returned to me 4 rows of NULLs in a numeric column that housed cubic feet per second values. This sparked a big conversation surrounding how to represent missing values in different data type columns. (NULL is a character type in R and creates the problem of having multiple data types in 1 column for numeric and datetime columns) We ended up deciding on NULL for missing values in character and categorical columns and NA for missing values in numeric and datetime columns, since NA is neither a character or numeric type in R. Once I changed the NULLs in numeric and datetime columns to NAs it ran through DPchecker just fine."