Frequency Tables for Continuous Variables

Discuss the jamovi platform, possible improvements, etc.
Post Reply
Frank.M.LoSchiavo
Posts: 20
Joined: Tue Apr 06, 2021 6:35 pm

Frequency Tables for Continuous Variables

Post by Frank.M.LoSchiavo »

I'm new to Jamovi (and R), but I really like Jamovi as an alternative to SPSS for my intro stat students. I sincerely thank the developers for their hard work! Quick issue...Using the exploration menu, I cannot find a way to create a frequency table when variables are continuous. I have found workarounds by using the Rj Editor and by transforming a continuous variable into a new nominal/ordinal variable with the same values. I really wish it were easier to create frequency tables no matter the measurement type, because there are several good reasons to create frequency tables regardless of the data. Simple example...I often create frequency tables as I begin cleaning my data. Is there any possibility of loosening the restrictions on frequency tables? Thanks again for your amazing work! - Frank
sbalci
Posts: 124
Joined: Sat Jan 06, 2018 10:25 pm
Contact:

Re: Frequency Tables for Continuous Variables

Post by sbalci »

you may use tableone function in ClinicoPath Descriptives Module
User avatar
MAgojam
Posts: 421
Joined: Thu Jun 08, 2017 2:33 pm
Location: Parma (Italy)

Re: Frequency Tables for Continuous Variables

Post by MAgojam »

Hi @Frank,
if you want to get a frequency table from the continuous variable, you can select it and from the ribbon Data>Setup>Measurement type classify it as Ordinal.
At this point with Analysis>Exploration>Descriptive and selecting the Frequency Tables checkbox, you will also get a frequency table of the values of your variable in the Variables list.
This works with both jamovi desktop and jamovi cloud.
If you use jamovi desktop, you can also follow the suggestion of Sbalci, who developed the indicated module and just preceded me in the suggestion.

Sorry if that's not what you were asking.

Cheers,
Maurizio
Frank.M.LoSchiavo
Posts: 20
Joined: Tue Apr 06, 2021 6:35 pm

Re: Frequency Tables for Continuous Variables

Post by Frank.M.LoSchiavo »

Sbalci, thanks for the module suggestion!

Maurizio, thanks for your reply, as well, although changing the measurement type doesn't fix the overall problem. I'm using jamovi desktop. If a variable's measurement type is classified as ordinal, I can produce a frequency table, but later, when I reclassify that same variable's measurement type to continuous, the frequency table will disappear! I realize that I can copy the frequency table and paste it somewhere for safekeeping, but it would be helpful to have all the output for all the analyses together in the Results window.

Perhaps I should state the issue another way...Using the same continuous variable, I'd like to have a frequency table (which currently is only available when measurement type is nominal/ordinal) and other analyses that require continuous variables (e.g., t-tests) within the same Results window. Bottom line...Why limit frequency tables to nominal/ordinal variables? Can that restriction be lifted in the base product?

Thx.

FML
User avatar
seol
Posts: 157
Joined: Mon Dec 04, 2017 8:40 am
Location: Seoul, Korea
Contact:

Re: Frequency Tables for Continuous Variables

Post by seol »

Hi @Frank,

why don't you create the existing var. into new variable as a nominal/ordinal variable shown below? you don't need to copy the frequency table and paste it somewhere.
Attachments
image.png
image.png (6.05 KiB) Viewed 22698 times
User avatar
MAgojam
Posts: 421
Joined: Thu Jun 08, 2017 2:33 pm
Location: Parma (Italy)

Re: Frequency Tables for Continuous Variables

Post by MAgojam »

Frank.M.LoSchiavo wrote: although changing the measurement type doesn't fix the overall problem.
Hi FML,
you are right and sorry for my simple answer.
If you are using the jamovi desktop version, I think you might find it useful to install the "PARALLEL USE OF STATISTICAL PACKAGES IN TEACHING" Rosetta module.
Take a look at the screenshot that I am attaching to you.
Campture.PNG
Campture.PNG (88.85 KiB) Viewed 22698 times
Cheers,
Maurizio
Frank.M.LoSchiavo
Posts: 20
Joined: Tue Apr 06, 2021 6:35 pm

Re: Frequency Tables for Continuous Variables

Post by Frank.M.LoSchiavo »

Seol, creating a new variable (which is continuous) is one of the workarounds I mentioned in my original post. At the moment, I think that's the best option without having to load additional modules.

Maurizio, thanks for the module suggestion. I'll definitely check it out!

I'm still hoping someone will address why the jamovi base limits frequency tables to nominal/ordinal variables, and if that restriction can be lifted in a future version.

Thx.

FML
User avatar
jonathon
Posts: 2609
Joined: Fri Jan 27, 2017 10:04 am

Re: Frequency Tables for Continuous Variables

Post by jonathon »

descriptives in jamovi is designed around the most common scenarios ... most of the time, you don't want frequency tables for continuous variables, because you'll end up with a giant table with as many rows in the table as there are rows in the data set.

normally we visualize a continuous variable with a histogram or similar.

cheers

jonathon
jmatthews
Posts: 12
Joined: Thu Jun 25, 2020 9:40 pm

Re: Frequency Tables for Continuous Variables

Post by jmatthews »

I'm with Frank on this one since it is somewhat challenging to sleuth out the modes in a distribution that has multiple modes (or close modes). The duplicate variable as ordinal is a workaround which I will teach in class but having a dedicated function for this would be nice. Just my two cents. Thanks!
eigenvalue
Posts: 4
Joined: Sat Aug 08, 2020 8:18 am

Re: Frequency Tables for Continuous Variables

Post by eigenvalue »

I'm bumping this request, as I agree with the point of view that continuous scored variables are commonly examined via frequency tables.

Example 1: What percentage of people earn $0 per week? In this example (real data), the answer is 13.4%. This is important information to describe/understand the data.

Example 2: A non trivial number of ability tests do not have that many items (fewer than 20), therefore, the corresponding frequency table is not that large. Moreover, when expressed as a percentage of questions answered correctly, researchers can determine the percentage of people who passed the test via a frequency table (i.e., achieved 50% or greater).

I could probably think of other examples, but these two alone are fairly compelling.

Thanks for re-considering this request :grin:
Attachments
frequency_table1.PNG
frequency_table1.PNG (19.04 KiB) Viewed 7227 times
frequency_table2.PNG
frequency_table2.PNG (20.9 KiB) Viewed 7227 times
Post Reply