(Having said that I have committed the ultimate sin of teaching SPSS to undergraduates. I encourage my MSc students to use something else)
-
-
Replying to @hisotalus
Spss is perfectly fine for most quanitative statistic tests. If I need to do basic t tests, anova or different types of regression I prefer the output in spss to matlab for example. I think avoiding hyperbole like ‘ultimate sin’ or pointless past graduation are a bit far
3 replies 0 retweets 2 likes -
Replying to @WilfNelson
I genuinely do not think it is a useful software to teach to undergraduates as a first approach to doing stats. If you like black box magic even then there are better performing OA software such as JASP. But we are gonna have to agree to disagree on this one.
2 replies 1 retweet 6 likes -
Replying to @hisotalus
Spss isn’t a black box, it can be but that is why we give statistics lectures. We spend years familiarising our students with the theory of statistics and the practice of using a stats calculator but theory is always first.
1 reply 0 retweets 1 like -
Replying to @WilfNelson
Sorry, I do not mean to be condescending but SPSS most definitely is a black box software. They do have a command line but it is very rare anyone who can run stats on SPSS command line would decide to do that as they are then almost certainly also capable of using R
4 replies 1 retweet 7 likes -
Replying to @hisotalus @WilfNelson
I think this is a case of your personal experience being applied too broadly. I was taught SPSS in my undergraduate and we were very much all taught to use the command line (and 3rd party toolboxes).
1 reply 0 retweets 0 likes -
SPSS jeopardises research because it's essentially not verifiable, not replicable, nor open source.
2 replies 2 retweets 12 likes -
Command line doesn't directly imply it's good but it's nice to hear you were taught CLI. That's rare. It sadly doesn't change the fact you should move to something open in an ideal world.
1 reply 1 retweet 1 like -
Agree fully about the ideal world. Not much time or money is set aside for ECRs to learn to do the things they already do, but better/more open/etc. How many people teach themselves Python/R during work hours vs after 6pm?
1 reply 0 retweets 0 likes -
I am not sure I agree with this as a true generalization across the board, but some people you are right. In the US, for example, you do get this training through out your graduate studies. And even in the UK, a lot of my peers and myself — we invest a LOT of time in expanding
2 replies 0 retweets 1 like
our skillset. A lot. We have little time but we do invest it wisely.
-
-
I think this is a poor argument - a part of being a good researcher is to harvest the essential research skills. Knowing how to analyze data is a part of that.
1 reply 0 retweets 2 likes - 3 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.