Suggestions for improvement

Hi,

I downloaded the trial version recently and I really like the program in general, however some features make the usage awkward for my work.

1. Missing values are not accepted.
Data frequently includes missing cells. These cannot just be deleted from the spreadsheet, nor can they be replaced by zero values whithout manipulating data contents. It would be helpful if StatistiXL could handle missing values like other programs.

2. Selection of data range
It would be very helpful if colums coul be selected which are not neighbouring. In a large dataset I would like to compare multiple colums (e.g. by correlation) at a time which are not neighbouring. Such selection is possible in Excel.

3. Spearman correlation
It would be nice to select more than just two columns. Other packages can do this.

4. Test for normal distribution
It is awkward that frequency data is required. Other packages transform the data accordingly in one step. In StatistiXL I have to go through frequency distribution first. This again is awkward because zero frequency cells are required at the beginning and the end. This is not done automaticly. The whole proceedure could be improoved a lot.

All other features are very good. Usage is very intuitive and easy to handle (better than in many other programs). Exact p values are calculated in nonparametric tests, very nice!

I will definitely watch further development of the program and buy it once it meets my demands.

Sincerely,

Dr. V. Witte, LMU, Munich

Comments

  • Hi Borriss

    Phil who is our statistician is away this week but I will ask him to have a look at your suggestions as soon as he returns. In brief however, I know we have discussed the issues of missing values before. While they are a common issue in real world datasets, the handling of them is problematic. We considered including missing value analysis in Version 1 of statistiXL and decided against it. Automatically handling missing values gives the false impression of it being a straightforward procedure to replace a missing value with an estimate. This is far from the case and in most instances, the appropriate values depend on knowledge that the researcher possesses and are not necessarily simple means etc that many packages implement. We are considering having a separate module in Version 2 to estimate missing values, which users could then use in subsequent analyses (at their own discretion). But we don't want the replacement of missing values to be a transparent process that the user might be unaware of.

    Point 2 is already supported. If you hole the CTRL key down you will be able to use the mouse to select non adjacent cells.

    Points 3 and 4 are both on the list of features that we plan to implement in future releases.

    King Regards

    Alan

  • Hi Alan,

    thanks for your quick reply.

    "Point 2 is already supported. If you hole the CTRL key down you will be able to use the mouse to select non adjacent cells"

    Selection in Excel works, but the data is not accepted by StatistiXL. I tried with Kruskal Wallis Test and Pearson Correlations.

    Another improvement would be in my opinion to leave all fields unchecked (or checked) by default in the data selection dialog boxes (after having used the same test before). In my version, when I use a method repeatedly, fields are somehow in between of being checked or unchecked (grayed) so that I have to go through all the settings again, even those I didn't use before.

    Best regards,

    Volker Witte
  • Hi Volker

    Your definitely seem to be having some unusual issues. I have just run the KW and Correlation stats on my system with disjoint ranges without issue. Using Example 2 on our Correlation spreadsheet (included with statistiXL) I entered the following as the data range for the simple correlation test

    Variable Range = Ex2 Multivariate!$A$1:$B$16,$D$1:$E$16

    Can you confirm that this doesn't work on your system?

    As to your second issue, statistiXL should simply remember the settings that you last used (i.e. if an option was checked the first time you run a test it should be checked the second time etc). I have had no previous reports of 1/2 way options. Can you please let me know the versions of Excel and Windows that you are running?

    Cheers

    Alan
  • Hi Alan,

    I am using Win XP Home, Version 2002, SP2 with Excel 2000, (9.0.2812). It is still the same, also with the example data, even if I copy your data selection range. The message is: invalid data range.

    Also the remembering of settings does not work correctly.

    I'll try a second computer at work

    Cheers,

    Volker
  • Hi Volker

    Please see the email I have sent you.

    Cheers

    Alan
Sign In or Register to comment.