search

IDEMSInternational / R-Instat

A statistics software package powered by R
http://r-instat.org/
GNU General Public License v3.0
38 stars 103 forks source link

Improving row statistics #4141

Open rdstern opened 7 years ago

rdstern commented 7 years ago

Prepare > Column: Calculate > Row Summary calculates row statistics, i.e. it produces a new column with (say) the mean of existing columns.

  1. It should have the option on what to do with missing values. Currently if any values are missing then the summary is missing.
  2. Like the ordinary calculations we should check what it does when a filter is applied.
  3. There is a disabled button to define a user function. Should that be added?
  4. Can we consider weighted row summaries?
shadrackkibet commented 6 years ago

It should have the option on what to do with missing values. Currently if any values are missing then the summary is missing.

@rdstern I see we can add na.rm to the apply function. Should we have a checkbox somewhere with the label Ignore Missing Values by default unchecked?

rdstern commented 6 years ago

That's right. We have that option on some other dialogues, e.g. Prepare > Column: Reshape > Summarise. It could usefully be consistent with that.

dannyparsons commented 6 years ago

Yes I needed this the other day!

rdstern commented 6 years ago

@shadrackkibet can this be closed?

shadrackkibet commented 6 years ago

2,3 and 4 are not yet sorted out. So, this cannot be closed.

dannyparsons commented 6 years ago

I think 1 was the most urgent so this can move to PENDING.

shadrackkibet commented 4 years ago

It should have the option on what to do with missing values. Currently if any values are missing then the summary is missing.

I think we can make the missing option more robust here by allowing the user to define missing value options like what we do in climatic summaries and the other ordinary summaries dialog. I needed this facility for some staff working at KMD who are using this dialog. I think it is now time to enable this feature.