toad.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
Mastodon server operated by David Troy, a tech pioneer and investigative journalist addressing threats to democracy. Thoughtful participation and discussion welcome.

Administered by:

Server stats:

273
active users

#spss

0 posts0 participants0 posts today

I really avoid ragging on #SPSS because for many things, it's a perfectly fine tool and I hate software wars. But while helping a colleague with some analyses conducted in SPSS, I noticed that the REGRESSION command treats predictors that are explicitly declared to be nominal variables still as scale variables (also without issuing any warning). That is really quite dumb. Of course users should notice that a single slope coefficient cannot represent a 5-level categorical variable, but still ...

Ich: Hier, lecker #Forschungsdaten und Dokumentation in offenen Formaten unter freier Lizenz! :awesome:

Deutsches Forschungsdatenzentrum: Wir haben Ihre Daten ins proprietäre #SPSS-Format konvertiert, weil das Programm sehr weit verbreitet ist. (Spoiler: Ist es gar nicht mehr, aber gut. Die Dokumentation ist damit jedenfalls größtenteils für die Tonne, weil für #Rstats und nicht für SPSS.)

Recently I asked for the data and code for a study that a journal invited me to review. Authors provided data, but "No code was used as all analysis were performed with SPSS."🤔
Let's see whether this is an exception to the rule that "It is often impossible to reproduce the results without knowing all the steps followed in the analysis, even when a detailed methodology is provided.", as this article found:
doi.org/10.1016/j.lansea.2024.
#peerreview #openscience #spss #reproducibility #verification

This week we're introducing our Special Interest Group leads! The @RoSE_Network_ has several Special Interest Groups (SIGs), each covering a different area of statistics education research.

We're starting with introducing our Statistics Software SIG! This SIG is interested in exploring questions about the educational pros and cons of different statistics software and how to teach and learn them effectively.

Replied in thread

@Brendanjones
They are only a problem when they are conducted within the atmosphere of this planet.

When O'Neill wrote about the #High #Frontier in 1970-80s the most obvious problems with colonies and #SPSS solar power satellites was what to do with/how to sell the electricity.

Move the bulk of heavy computation off the planet, power it from solar satellites - probably photovoltaic, but perhaps thermal systems may have places - and move results and givens up and down, economically of power.

Replied in thread

@r_ivorra I don't get the impression that #rstats is declining. There seem to be some people using #julialang instead of (or in addition to) R, but there are also people picking up R instead of packages like #spss.

As for competition from #python, I'm not sure how significant this is. Among those of my classmates who are inexperienced programmers, there is some strong R-avoidance in favour of Python; those people were never going to consider R anyway.