Spss

[joshie6891]

This seems like a good place to ask this as any.

I'm working on a project with my professor right now, and she emailed me the data outputs from our meeting earlier today.

Is there any way to me to view the output without a copy of SPSS? I'm an undergraduate student, so it's not really worth it for me to have a copy since I only use it sporadically (aaah, $$$$), but it would be great for me to be able to mull over the data while at home on my own computer.

Thanks!

-J

---

Members don't see this ad.

 

[Therapist4Chnge]

I was able to get a copy from my undergrad (free) IT dept, and I think they had discount copies to purchase. It may be worth inquiring about a copy, you may be able to get it for $10-$25. I know M$ sometimes has similar deals for their products.

-t

 

[phd2006]

If you can, purchase SPSS (and may be inexpensive through your school as T4C indicated) - you never know when you may need to work with the actual data yourself.

If you're not going to purchase SPSS, have your professor export the outputs as Word documents (File > Export> choose Word .doc as the file export type). The formatting is a little off in the Word doc, but you'll have all the information from the output file.

 

[psychwanabe]

Can't you also export to Excel? That might be a little more consistent.

Also, the student version of SPSS can be had for $100. You only get a 4 year license with it but if you prefer to work at home it's worth the money.

 

[Therapist4Chnge]

My copy died with one of my HD failures, but it was REALLY helpful to have around, especially when it was getting late in the computer lab....I could just go home and finish. There are few things that are more fun than playing with SPSS in your pajamas!!

-t

 

[joshie6891]

Hm. Thanks for all the advice, everyone! The cheapest version I could find through my school was $200, which is a little too pricey for now. I'm planning on taking a bit of time off, so it's probably not worth buying right away.

I ended up downloading the 15 day trial. That will get me through for now, and if I really end up needing it later, I can reconsider.

 

[phd2006]

Can't you also export to Excel? That might be a little more consistent...

Yes. You can export to a number of formats - word, excel, html, pdf, powerpoint, text, etc. I suggested Word as most people I work with are more comfortable manipulating the output tables in Word for presentations/papers (I use many formats for exporting). Most likely depends on your purposes & your comfort level with the other formats.

 

[psychwanabe]

There are few things that are more fun than playing with SPSS in your pajamas!!

Okay, so that just made me snort with laughter and now I feel very goofy here in my computer lab! Only a true psych nerd would say that, and only another would recognize it as true!!!

Hm. Thanks for all the advice, everyone! The cheapest version I could find through my school was $200, which is a little too pricey for now. I'm planning on taking a bit of time off, so it's probably not worth buying right away.

I ended up downloading the 15 day trial. That will get me through for now, and if I really end up needing it later, I can reconsider.

Joshie if you decide to buy, just google SPSS and you'll find the mfgr's website (Prentice Hall I think). From there, look for the student version and you can order it online. It's quick and easy. It was even a few bucks cheaper than buying it from a store!

 

[brightness]

I hate statistics. You are all on high on drugs. No, but really, its just the hand calculations I hate. So I do love SPSS, but only because it saves me from doing math by hand.

Okay, so that just made me snort with laughter and now I feel very goofy here in my computer lab! Only a true psych nerd would say that, and only another would recognize it as true!!!



Joshie if you decide to buy, just google SPSS and you'll find the mfgr's website (Prentice Hall I think). From there, look for the student version and you can order it online. It's quick and easy. It was even a few bucks cheaper than buying it from a store!

 

[JockNerd]

I hate statistics. You are all on high on drugs. No, but really, its just the hand calculations I hate. So I do love SPSS, but only because it saves me from doing math by hand.

Oh, the fun only BEGINS when you learn the formulas and know what SPSS is doing behind the scenes! There are some hand calcs you can't get away from anyhow (some effect sizes and such). But, really, the interesting parts come when you know WHY you're getting those stats, and can see the relationships in your data in your mind.

I love stats and research

The Student version is pretty limited, Joshie. You'd be able to view data and carry out some simple calculations, but I think there's a limit of, what, 250 entries? I'd just use a lab/library computer to look at the data if I were you.

 

[Therapist4Chnge]

You should get the other grad students together and chip in for a futon, some pillows, maybe one of those tooth brush holder things....ya know, just so you are prepared.

1 black futon......$299
6 pillow..........$48
1 tooth brush holder......$2.99
Being able to sleep next to your data for 4-5 years......PRICELESS!!

-t

 

[psychwanabe]

Oh, the fun only BEGINS when you learn the formulas and know what SPSS is doing behind the scenes! There are some hand calcs you can't get away from anyhow (some effect sizes and such). But, really, the interesting parts come when you know WHY you're getting those stats, and can see the relationships in your data in your mind.

I love stats and research

The Student version is pretty limited, Joshie. You'd be able to view data and carry out some simple calculations, but I think there's a limit of, what, 250 entries? I'd just use a lab/library computer to look at the data if I were you.

Well...I'm not completely sure about that. I just bought 15.0 and it says it can't open data files with more than 50 variables or 1500 cases. That seems okay to me unless I'm missing something. I can't tell you for sure, because I haven't installed it yet. I'm holding out as long as I can to stretch my four years out!!

As far as loving stats I am very excited to finally be learning all the freaky-fun things I can do with SPSS. It's one of the highlights of my week, I gotta say.

You should get the other grad students together and chip in for a futon, some pillows, maybe one of those tooth brush holder things....ya know, just so you are prepared.

1 black futon......$299
6 pillow..........$48
1 tooth brush holder......$2.99
Being able to sleep next to your data for 4-5 years......PRICELESS!!

-t

Yeah, I snorted again. You are embarrasing me.

 

[Ollie123]

Check with the school. It was 90 bucks for the full version at my old school, and 10 bucks for the full version here. Either is better than the crap-ass student version that I believe doesn't contain all tests, and definitely has the limits psychwanabe specified.

If you can work with 50 variables than its a good choice, but most of the stuff I see has way wayyyyyy more than that. Heck we use some single measures with more than 50 items. I guess you could split files and do partial scorings and then combine them but blech. That sounds like its asking for trouble.

Every major university I've heard of offers serious discounts on it, so I'd check. If you can get a full version its well worth.

I'm also jealous you all get to learn SPSS. We do SAS here. More powerful program, but a LOT more archaic. I'm not a fan.

 

[Therapist4Chnge]

I have yet to play with SAS. I was curious how prevalent it is, compared to SPSS. I can't believe it is up to v15.....I think I originally learned on SPSS 8.

-t

 

[paramour]

The student version typically costs around $90 and yes it is absolutely craptastic. I wouldn't recommend it unless you are perhaps just starting out in stats and only need it for homework assignments or something simple. I eventually ended up w/ the grad student version for $200 and it included a heck of a lot more w/o the severe limitations of the base student version.

And I positively lurvs SAS! I miss it though b/c it appears most people around here work with SPSS. Stats prof in my master's program required SAS instead of SPSS b/c he claimed it was more useful than SPSS, you could easily go from SAS to SPSS (rather than vice versa), and "if you work in the *real* world, particularly w/ gov't positions, then you'll need to know it." I recall asking about SPSS vs. SAS at places I interviewed and most people thought I was nuts when I asked about SAS. It does appear to be a useful thing to know, however, as every now & again, I see job postings for people w/ SAS experience starting out at $75K+... for JUST SAS. I suppose my prof wasn't lying when he said he made decent money doing stats consultant work.

 

[brightness]

See, I so want to do all the clinical activities of a psychologist, but I'm not super interested in conducting research. I find it interesting, and I feel like I'll do what I have to do, but I don't want to make research a real focus of my career.
Ok, sorry, I'm off topic. To the OP, I usually just print the data set off before I come home in a lab on campus. It only takes a couple seconds and costs .10, so for me it is worth it.

Oh, the fun only BEGINS when you learn the formulas and know what SPSS is doing behind the scenes! There are some hand calcs you can't get away from anyhow (some effect sizes and such). But, really, the interesting parts come when you know WHY you're getting those stats, and can see the relationships in your data in your mind.

I love stats and research

The Student version is pretty limited, Joshie. You'd be able to view data and carry out some simple calculations, but I think there's a limit of, what, 250 entries? I'd just use a lab/library computer to look at the data if I were you.

 

[Therapist4Chnge]

[off topic]If you can speak in business terms and make it relevant to the firm/company = $$$$. Also, if you can make pretty graphs and charts you can tack on another few grand (not kidding). I strongly suggest consulting contracts and not salaried positions.[/off topic]

-t

 

[Ollie123]

Actually one of the biggest problems with SAS is it produces the most horrific looking graphs you will ever see. Visually, its about at the level of a 1980's DOS program.

Paramour does have a point about it being alot easier to go from SAS to SPSS though. SAS (at least here) is pure syntax so no point and click. Means you've really gotta know your stuff to get it working. My opinion only, but I think everyone should be required to learn syntax, though apparently some schools don't teach it.

I'm shooting for a salaried position at a university with an opportunity to consult on the side doing nerdier things (psychophys setups, software programming, stats, etc.). More stability and I can still get a decent income boost.

 

[Therapist4Chnge]

Old versions of SPSS were more syntax heavy, now everything is pretty much point and click with some manual steps here and there (at least for what I used it for, I'm outta heavy stats crunching now...I know enough to let the stat lovers tweak stuff)

I once did a data dump from some SQ tables to excel, and then played around with it and came up with some colorful graphs, and the people I presented to couldn't get past how cool it looked. They didn't even care I isolated their problem and re-designed their implementation method, they just wanted to know how I made the graphs 'from a bunch of numbers'. I knew at that moment that I always had a job in Corp America making pretty graphs for high level execs to show to their colleagues. The best part about the whole deal was that the guy who loved it the most was the #2 at said billion dollar company, and he appointed me to a committee with my boss' boss' boss.....talk about a priceless reaction when she saw me enter the room with him. She didn't really like me to begin with, but he bought me all sorts of leverage after that.

-t

 

[psychanon]

You can do anything in SPSS in either syntax or point and click (except for macros, which you can just do in syntax). Syntax is always the way to go-- because it can be saved, checked for accuracy, shared with colleagues, re-run, etc. etc.

My university offers the full version for $50, but note that this is just for one year-- after that, you have to get a new license (a sucky new system in my opinion).

50 variables is really not enough. I can't believe they'd set that low a limit. To give you some contexts, I just checked the main database for one of my lab studies that i use for my analyses-- it has 4,727 variables. About 4300 are from the study itself (it was longitudinal), the remaining 400 are variables I've computed for my analyses.

I just got the new SPSS 15. They changed the look of it a lot. They also made it Java-based, which makes it freeze up and generally suck. I also hate how you can't read the output that you made from an earlier version. But if you want to do MLM in SPSS, you need at least 14.

I'm with many of you-- I love doing data analyses. I think it's so exciting! The statistics part itself bores me-- reading a statistics article is about the dullest thing I can imagine--- but once I know what I'm doing, I love examining data. Such nerds are we....

 

[Therapist4Chnge]

I just got the new SPSS 15. They changed the look of it a lot. They also made it Java-based, which makes it freeze up and generally suck.

Oh god. 90% of java programs are junk. It is not the friendliest language to program, and most programmers just aren't that great with it.....which leads to sluggish/inconsistent performance. From what I recall about java programming, it isn't a big deal for small apps, but when you have complex apps with heavy data getting thrown around, I can't see it doing well. Maybe it is different with faster processors, but back when I learned Java (maybe 1998 or 1999?) it would freeze up the computer and do all sorts of other weird stuff when I tried to put any stress on a program (running a ton of loops and whatnot to see how it'd react). They should have stuck with the DOS roots and used a different language. We had to design a scientific calculator program and I added a function on there that would keep dividing the number over and over again (sorta like asking the computer to find PI to the nth number).....it worked, but it would crash the system after not too long.

-t

 

[psychwanabe]

hmmm...well, maybe I won't be happy with it once I load it on. It was my stats prof who recommended it though, so I can blame him if I'm not happy!

 

[Therapist4Chnge]

hmmm...well, maybe I won't be happy with it once I load it on. It was my stats prof who recommended it though, so I can blame him if I'm not happy!

Take my comments with a grain of salt, I'm not a programmer, and I'm not up to date with programming stuff....I just have a bias against java programming because I haven't seen many that do it well. For all I know it works great.

Anyone use 15 yet?

-t

 

[phd2006]

Don't buy the Student Version -- as others have pointed out, too limiting. At least get the Graduate Student Version -- it won't have MVA and a few other cool features, but it will have most of the analyses you need.

So far I haven't had any problems with version 15 opening data, syntax and output files from earlier versions (versions 11 through 14) -- the graphics are sometimes messed up, but all the tables have opened. Then again, I started using SPSS so many versions ago -- when it was all syntax, no pointing & clicking -- so I'm pretty used to the upgrades not always being 100% compatible with earlier versions.

If you have the opportunity, learn both SPSS and SAS (as well as at least one structural equation modeling program) -- it will greatly increase your marketability. Even just knowing how to do most analysis in Excel is a bonus to many.

Also, as others have said -- learn syntax! I can't believe this isn't being taught more frequently!! It's a record of what you did -- how will you remember 6 months later how you created that scale average (among the hundreds of other processes you will have done)?

And graphing -- it's all about data representation. Devleop these skills!! One stat elective (yes, I'm that person, took stats electives) that has been extremely beneficial was an advanced graphing course.

 

[JockNerd]

If you have the opportunity, learn both SPSS and SAS (as well as at least one structural equation modeling program) -- it will greatly increase your marketability. Even just knowing how to do most analysis in Excel is a bonus to many.

I agree 100%

 

[paramour]

For those contemplating a SPSS purchase,

Available for Microsoft® Windows® and Macintosh®, the SPSS Graduate Pack includes:

• Full version of SPSS Base
• Two add-on modules, SPSS Regression™ and SPSS Advanced Models™
• Software for structural equation modeling, Amos™ (available only on Windows)

As your students' analytical needs increase, the SPSS Graduate Pack grows right along with them. They can purchase additional add-on modules and software for specialized techniques, such as complex sampling and correspondence analysis, publication-ready tabular reporting, and much more.

Best of all, the SPSS Graduate Pack is affordable for students—up to 85 percent off * the commercial list price of SPSS Base.

http://www.spss.com/gradpack/

 

[paramour]

Actually one of the biggest problems with SAS is it produces the most horrific looking graphs you will ever see. Visually, its about at the level of a 1980's DOS program.

Paramour does have a point about it being alot easier to go from SAS to SPSS though. SAS (at least here) is pure syntax so no point and click. Means you've really gotta know your stuff to get it working. My opinion only, but I think everyone should be required to learn syntax, though apparently some schools don't teach it.

I'm shooting for a salaried position at a university with an opportunity to consult on the side doing nerdier things (psychophys setups, software programming, stats, etc.). More stability and I can still get a decent income boost.

I agree--SAS does have some awful graphing options. We switched versions between semesters and the newer one had upgraded to some niftier graphing capabilities but we still produced better graphs by simply exporting the info & making them elsewhere.

I am aware that you can use syntax for most functions in SPSS, but I know very few people (mostly stats profs) who are aware of and actually use them. It seems that most think it's enough of a struggle getting students to take stats and then utilize it, much less forcing them learn a few strands of syntax.

 

[psychwanabe]

Also, as others have said -- learn syntax! I can't believe this isn't being taught more frequently!! It's a record of what you did -- how will you remember 6 months later how you created that scale average (among the hundreds of other processes you will have done)?

We went over L Matrices today in stats. All I can say is that I need practice! I am NOT looking forward to that exam.

My prof decided to show us how good we had it and he brought up his stats from his dissertation (pre-SPSS). I can't even imagine having to write all that for statistical analysis and then having it error out because of one tiny mistake you then have to go dig for.

God bless technology!!

 

[psychanon]

So far I haven't had any problems with version 15 opening data, syntax and output files from earlier versions (versions 11 through 14) -- the graphics are sometimes messed up, but all the tables have opened. Then again, I started using SPSS so many versions ago -- when it was all syntax, no pointing & clicking -- so I'm pretty used to the upgrades not always being 100% compatible with earlier versions.

Really? I haven't had any problems with syntax or data files, but I haven't been able to open the output (I guess if you're going to have problems with one kind of file, it's best to have it be output, but it's still annoying). In fact, it seems that the extension has changed for the output files-- from .spo (output) to .spv ("viewer" files). Any idea what I could be doing wrong? I made a pretty big jump-- from 12 to 15.

 

[JockNerd]

I hate statistics. You are all on high on drugs. No, but really, its just the hand calculations I hate. So I do love SPSS, but only because it saves me from doing math by hand.

Just to illustrate, I'm hand-calculating t-tests right now. Gotta test for differences in two administrations of a test, and I only have the means and SDs from one of the administrations. It's really not so bad.

 

[psychwanabe]

I hate statistics. You are all on high on drugs. No, but really, its just the hand calculations I hate. So I do love SPSS, but only because it saves me from doing math by hand.

Maybe I'm weird, but I took a real comfort in doing stuff by hand. It's all organized and structured...just follow the rules and you are good! I loved it.

...would that all of life were that way!

 

[Therapist4Chnge]

Maybe I'm weird, but I took a real comfort in doing stuff by hand. It's all organized and structured...just follow the rules and you are good! I loved it.

...would that all of life were that way!

I preferred it this way in the beginning, because I could check my calculations (3x....not always, but at least 2x!). The problem was it took forever, but I felt comfortable with it. When SPSS switched around some stuff (I think it was 8-->10) it irked me.

Change...I love change...or not.

-t

 

[razzled]

Also, "R" is free and considered to be more sophisticated than SPSS...once you learn to use it, it's totally worth it to have. You can just google CRAN and download it...

 

[phd2006]

...My prof decided to show us how good we had it and he brought up his stats from his dissertation (pre-SPSS). I can't even imagine having to write all that for statistical analysis and then having it error out because of one tiny mistake you then have to go dig for. ...God bless technology!!

So true. It so amazes me when I hear/read about students doing scale reliability (not cross validation or scale development) or scale group differences (on established measures) as their dissertation. That was my data screening!! Technology allows us to do so much more.

 

[phd2006]

Really? I haven't had any problems with syntax or data files, but I haven't been able to open the output (I guess if you're going to have problems with one kind of file, it's best to have it be output, but it's still annoying). In fact, it seems that the extension has changed for the output files-- from .spo (output) to .spv ("viewer" files). Any idea what I could be doing wrong? I made a pretty big jump-- from 12 to 15.

IIRC, there was a big jump from version 11 to version 12 (or 13). (IIRC there were also incompatibility issues between version 8/9 and v10). Output files prior to version 12 (or 13) first need to be converted to version 12 (or 13) to be opened in versions 14 and 15.

If you jumped from v12 to v15, I'm suspect that you 'missed' a step. That is, each time SPSS upgrades, you may need to make sure that files can open & function in each successive version. Don't 'skip' versions - each has to be 'converted' in succession.

I have version 11 on my home desktop (just haven't upgraded...although, it's looking like I should soon), which worked with converting files to version 14. I work on version 15 on campus and so far have no difficulties accessing at home in version 11 (note: all files have been upgraded/compatible with version 14). All is on XP operating system. I just got a new laptop with the Vista operating system, and have not yet had the opportunity to see if any incompatibility issues occur (although, I am suspecting some).

 

[joshie6891]

Wow, thanks for all the good information, everyone! At this point, it looks like the cheapest version is still too pricey for how much I'll be using the program in the near future. For now, I'll probably get by with the trial version. After our next experiment is over, I'll definitely have to go to the library for analysis.... but at least the trial gets me through this project's analysis