Can a correlation matrix be printed easily ?
Locked
 
|
…can SPSS print a simple correlation matrix/table without the default Sig and N rows ? In other words, just the correlations and nothing else? There’s no option in the correlate menu item windows to do this and I can’t find any option anywhere to suppress the unneeded output with any type of syntax option. I’m trying to keep this simple since I’m using an example for teaching purposes so I don’t want to use complex output manipulation syntax. Best, Jeff |
Re: Can a correlation matrix be printed easily ?
|
|
Jeff, One way to get this is to double-click the correlation table to put it into edit mode. Then, move the statistics to the layer dimension. You might need to use the Pivot menu to make the pivot tray visible. As I think you indicate in your note, OMS would be another way. Tony Babinec From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Jeff A …can SPSS print a simple correlation matrix/table without the default Sig and N rows ? In other words, just the correlations and nothing else? There’s no option in the correlate menu item windows to do this and I can’t find any option anywhere to suppress the unneeded output with any type of syntax option. I’m trying to keep this simple since I’m using an example for teaching purposes so I don’t want to use complex output manipulation syntax. Best, Jeff ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
Re: Can a correlation matrix be printed easily ?
Administrator
|
In reply to this post by Jeff A
I think FACTOR with /PRINT = CORRELATION will give you what you want.
RELIABILITY with /SUMMARY=CORR is another option. Both give plain text output, IIRC. HTH. Jeff-2 wrote > .can SPSS print a simple correlation matrix/table without the default Sig > and N rows ? In other words, just the correlations and nothing else? > > > > There's no option in the correlate menu item windows to do this and I > can't > find any option anywhere to suppress the unneeded output with any type of > syntax option. > > > > I'm trying to keep this simple since I'm using an example for teaching > purposes so I don't want to use complex output manipulation syntax. > > > > Best, > > > > Jeff > > > > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > (not to SPSSX-L), with no body text except the > command. To leave the list, send the command > SIGNOFF SPSSX-L > For a list of commands to manage subscriptions, send the command > INFO REFCARD ----- -- Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
Re: Can a correlation matrix be printed easily ?
|
|
Hello, A nifty trick I learned some time ago: Go to Scale, Reliability Analysis, Statistics and check correlations. Best correlation matrix in SPSS. Bob On Sun, Apr 7, 2019 at 7:18 AM Bruce Weaver <[hidden email]> wrote: I think FACTOR with /PRINT = CORRELATION will give you what you want. |
|
|
In reply to this post by Jeff A
In addition to the correlation matrices available in other procedures and using the pivot table editor, REGRESSION, which may be more familiar to your students, can give you the correlation matrix. It does include sections with the sig levels and N's, but the layout is such that the correlations all appear together, so it is easier to study them. And since this matrix is closely related to the coefficient correlations, it may be instructive to have them together. On Sun, Apr 7, 2019 at 4:51 AM Jeff A <[hidden email]> wrote:
|
|
|
In reply to this post by Bruce Weaver
Thanks Bruce, Robert, and Jon,
I sometimes wonder why SPSS doesn't provide more direct options to provide less-cluttered output. The reliability syntax will do what I need (although it gives unnecessary output). Jeff -----Original Message----- From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Bruce Weaver Sent: Monday, April 8, 2019 12:18 AM To: [hidden email] Subject: Re: Can a correlation matrix be printed easily ? I think FACTOR with /PRINT = CORRELATION will give you what you want. RELIABILITY with /SUMMARY=CORR is another option. Both give plain text output, IIRC. HTH. Jeff-2 wrote > .can SPSS print a simple correlation matrix/table without the default > Sig and N rows ? In other words, just the correlations and nothing else? > > > > There's no option in the correlate menu item windows to do this and I > can't find any option anywhere to suppress the unneeded output with > any type of syntax option. > > > > I'm trying to keep this simple since I'm using an example for teaching > purposes so I don't want to use complex output manipulation syntax. > > > > Best, > > > > Jeff > > > > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > (not to SPSSX-L), with no body text except the command. To leave the > list, send the command SIGNOFF SPSSX-L For a list of commands to > manage subscriptions, send the command INFO REFCARD ----- -- Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
Re: Can a correlation matrix be printed easily ?
|
Administrator
|
Jeff, FACTOR can give you a pivot table showing the correlation matrix with
very little other output. E.g., * Use FACTOR command to generate a correlation matrix with correlations only. * Change path on next line as needed. GET FILE='C:\SPSSdata\survey_sample.sav'. FACTOR /VARIABLES educ age paeduc maeduc /ANALYSIS educ age paeduc maeduc /PRINT CORRELATION /ROTATION NOROTATE . * Confirm via CORRELATIONS command with LISTWISE deletion. CORRELATIONS /VARIABLES=educ age paeduc maeduc /PRINT=TWOTAIL NOSIG /MISSING=LISTWISE. Jeff-2 wrote > Thanks Bruce, Robert, and Jon, > > I sometimes wonder why SPSS doesn't provide more direct options to provide > less-cluttered output. > > The reliability syntax will do what I need (although it gives unnecessary > output). > > Jeff ----- -- Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
Re: Can a correlation matrix be printed easily ?
|
Administrator
|
PS- You can omit the /ANALYSIS sub-command if you wish, and shorten the
command to: FACTOR /VARIABLES educ age paeduc maeduc /PRINT CORRELATION /ROTATION NOROTATE . Bruce Weaver wrote > Jeff, FACTOR can give you a pivot table showing the correlation matrix > with > very little other output. E.g., > > * Use FACTOR command to generate a correlation matrix with correlations > only. > * Change path on next line as needed. > GET FILE='C:\SPSSdata\survey_sample.sav'. > > FACTOR > /VARIABLES educ age paeduc maeduc > /ANALYSIS educ age paeduc maeduc > /PRINT CORRELATION > /ROTATION NOROTATE > . > * Confirm via CORRELATIONS command with LISTWISE deletion. > CORRELATIONS > /VARIABLES=educ age paeduc maeduc > /PRINT=TWOTAIL NOSIG > /MISSING=LISTWISE. > > > > > Jeff-2 wrote >> Thanks Bruce, Robert, and Jon, >> >> I sometimes wonder why SPSS doesn't provide more direct options to >> provide >> less-cluttered output. >> >> The reliability syntax will do what I need (although it gives unnecessary >> output). >> >> Jeff > > > > > > ----- > -- > Bruce Weaver > bweaver@ > http://sites.google.com/a/lakeheadu.ca/bweaver/ > > "When all else fails, RTFM." > > NOTE: My Hotmail account is not monitored regularly. > To send me an e-mail, please use the address shown above. > > -- > Sent from: http://spssx-discussion.1045642.n5.nabble.com/ > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > (not to SPSSX-L), with no body text except the > command. To leave the list, send the command > SIGNOFF SPSSX-L > For a list of commands to manage subscriptions, send the command > INFO REFCARD ----- -- Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
|
|
Yup, That one works even better. ...don't know why they don't just add an
option to the regular correlations command to suppress all of the extra output and just print this same simple table as you can do below. A correlation matrix is so common that several on this list have already figured out their own work-arounds. ...would even be better if spss offered an option to print only the lower half of the matrix. I love easy to read output. So much easier to read quickly, especially when you're scrolling up/down to make different comparisons. Jeff -----Original Message----- From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Bruce Weaver Sent: Tuesday, April 9, 2019 4:30 AM To: [hidden email] Subject: Re: Can a correlation matrix be printed easily ? PS- You can omit the /ANALYSIS sub-command if you wish, and shorten the command to: FACTOR /VARIABLES educ age paeduc maeduc /PRINT CORRELATION /ROTATION NOROTATE . Bruce Weaver wrote > Jeff, FACTOR can give you a pivot table showing the correlation matrix > with > very little other output. E.g., > > * Use FACTOR command to generate a correlation matrix with correlations > only. > * Change path on next line as needed. > GET FILE='C:\SPSSdata\survey_sample.sav'. > > FACTOR > /VARIABLES educ age paeduc maeduc > /ANALYSIS educ age paeduc maeduc > /PRINT CORRELATION > /ROTATION NOROTATE > . > * Confirm via CORRELATIONS command with LISTWISE deletion. > CORRELATIONS > /VARIABLES=educ age paeduc maeduc > /PRINT=TWOTAIL NOSIG > /MISSING=LISTWISE. > > > > > Jeff-2 wrote >> Thanks Bruce, Robert, and Jon, >> >> I sometimes wonder why SPSS doesn't provide more direct options to >> provide >> less-cluttered output. >> >> The reliability syntax will do what I need (although it gives unnecessary >> output). >> >> Jeff > > > > > > ----- > -- > Bruce Weaver > bweaver@ > http://sites.google.com/a/lakeheadu.ca/bweaver/ > > "When all else fails, RTFM." > > NOTE: My Hotmail account is not monitored regularly. > To send me an e-mail, please use the address shown above. > > -- > Sent from: http://spssx-discussion.1045642.n5.nabble.com/ > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > (not to SPSSX-L), with no body text except the > command. To leave the list, send the command > SIGNOFF SPSSX-L > For a list of commands to manage subscriptions, send the command > INFO REFCARD ----- -- Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
|
|
Statistics tends to give the output that is most commonly wanted and then provide editing facilities to adjust further such as the pivot table editor. In this case, pivoting does the job, and that can be automated with this syntax. SPSSINC MODIFY TABLES subtype="'Correlations'" SELECT=0 /STYLES CUSTOMFUNCTION="customstylefunctions.moveRowsToLayers(fromdim=0,todim=0)". I know you don't like these extension commands and post edits, but this extension command can enhance correlation matrices in a number of ways, including blanking the upper triangle, hiding insignificant correlations, and bolding large ones. On Mon, Apr 8, 2019 at 3:18 PM Jeff A <[hidden email]> wrote: Yup, That one works even better. ...don't know why they don't just add an |
|
|
Thanks Jon, Actually I like the extra extensions for myself, but my social-science/math-code-phobic students can’t deal with the complexity. I’m experimenting this year with illustrating a number of things using hypothetical data created with the “Make Cases” dialog you provided before and it seems to be working well so far, but that’s not for the students. Having them run their own correlation matrixes and having them run several regressions is working much better than trying to illustrate things like control variables in some other way, but I need to keep the syntax simple for them. Thanks for the assistance. Jeff From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Jon Peck Statistics tends to give the output that is most commonly wanted and then provide editing facilities to adjust further such as the pivot table editor. In this case, pivoting does the job, and that can be automated with this syntax. SPSSINC MODIFY TABLES subtype="'Correlations'" SELECT=0 /STYLES CUSTOMFUNCTION="customstylefunctions.moveRowsToLayers(fromdim=0,todim=0)". I know you don't like these extension commands and post edits, but this extension command can enhance correlation matrices in a number of ways, including blanking the upper triangle, hiding insignificant correlations, and bolding large ones. On Mon, Apr 8, 2019 at 3:18 PM Jeff A <[hidden email]> wrote:
-- Jon K Peck ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
Re: Can a correlation matrix be printed easily ?
|
Jeff Phobia I'm with you on social-science/math-code-phobic students. Such statistical knowledge and skill that I have is mostly self-taught "on-the-job", so, when teaching, I could not only empathise with such students, but also present complex statistical ideas in (sometimes idiosyncratic) ways which they could appreciate, understand and, most important, remember. Students in the social sciences are not interested in the standard deviation of the lengths of a lorry-load of planks, but they do want to know how to measure authoritarian attitudes and how such attitudes might relate to the way people vote, especially when you call a derived variable [redkneck]. Correlation The simple correlation matrix can also be eyeballed to create clusters of variables (not factors, but close). I used to get students to find the highest correlation in the matrix, circle it, and then look along the same row to find the next highest correlation and circle that. From this second correlation, look down the same column, find the highest correlation in the column and circle that. Continue alternately in horizontal and vertical columns until no other variable has its highest correlation with the last variable circled. The circled variables constitute the first cluster. The second and subsequent clusters are created using the remaining variables until all variables have been checked. Students can then discuss the nature of the resulting clusters. On the very last evening of the last post-grad class before I retired, I referred to some advanced statistics not covered in the course. One of these was cluster analysis, to illustrate which I told my students I knew a joke involving a spittoon, but which was too disgusting to repeat. Unanimously they shrieked, "Tell us!" and continued chanting until I did. Brought the house down, but at least they then knew what cluster analysis was. The joke has been around for many years. One version is on pp. 76-77 of: L Vine-Knight https://www.smashwords.com/profile/view/leovineknight Looking through the Windows of madness (2009) https://www.amazon.co.uk/gp/product/1601458827/ref=dbs_a_def_rwt_bibl_vppi_i0 As with all good pub jokes, longer versions (to infuriate listeners and risk physical attack by drinking companions) spin the story out by repeatedly increasing the amount of the wager until the victim accepts it. Enjoy! John F Hall [Retired academic survey researcher] IBM-SPSS Academic Author 9900074 Email: [hidden email] Website: http://surveyresearch.weebly.com/ SPSS course: http://surveyresearch.weebly.com/1-survey-analysis-workshop-spss.html Research: http://surveyresearch.weebly.com/3-subjective-social-indicators-quality-of-life.html From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Jeff A Thanks Jon, Actually I like the extra extensions for myself, but my social-science/math-code-phobic students can’t deal with the complexity. I’m experimenting this year with illustrating a number of things using hypothetical data created with the “Make Cases” dialog you provided before and it seems to be working well so far, but that’s not for the students. Having them run their own correlation matrixes and having them run several regressions is working much better than trying to illustrate things like control variables in some other way, but I need to keep the syntax simple for them. Thanks for the assistance. Jeff From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Jon Peck Statistics tends to give the output that is most commonly wanted and then provide editing facilities to adjust further such as the pivot table editor. In this case, pivoting does the job, and that can be automated with this syntax. SPSSINC MODIFY TABLES subtype="'Correlations'" SELECT=0 /STYLES CUSTOMFUNCTION="customstylefunctions.moveRowsToLayers(fromdim=0,todim=0)". I know you don't like these extension commands and post edits, but this extension command can enhance correlation matrices in a number of ways, including blanking the upper triangle, hiding insignificant correlations, and bolding large ones. On Mon, Apr 8, 2019 at 3:18 PM Jeff A <[hidden email]> wrote:
-- Jon K Peck ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
|
|
If I tell any more jokes to my students, I think they’ll try to hang me. J From: SPSSX(r) Discussion <[hidden email]> On Behalf Of John F Hall Jeff Phobia I'm with you on social-science/math-code-phobic students. Such statistical knowledge and skill that I have is mostly self-taught "on-the-job", so, when teaching, I could not only empathise with such students, but also present complex statistical ideas in (sometimes idiosyncratic) ways which they could appreciate, understand and, most important, remember. Students in the social sciences are not interested in the standard deviation of the lengths of a lorry-load of planks, but they do want to know how to measure authoritarian attitudes and how such attitudes might relate to the way people vote, especially when you call a derived variable [redkneck]. Correlation The simple correlation matrix can also be eyeballed to create clusters of variables (not factors, but close). I used to get students to find the highest correlation in the matrix, circle it, and then look along the same row to find the next highest correlation and circle that. From this second correlation, look down the same column, find the highest correlation in the column and circle that. Continue alternately in horizontal and vertical columns until no other variable has its highest correlation with the last variable circled. The circled variables constitute the first cluster. The second and subsequent clusters are created using the remaining variables until all variables have been checked. Students can then discuss the nature of the resulting clusters. On the very last evening of the last post-grad class before I retired, I referred to some advanced statistics not covered in the course. One of these was cluster analysis, to illustrate which I told my students I knew a joke involving a spittoon, but which was too disgusting to repeat. Unanimously they shrieked, "Tell us!" and continued chanting until I did. Brought the house down, but at least they then knew what cluster analysis was. The joke has been around for many years. One version is on pp. 76-77 of: L Vine-Knight https://www.smashwords.com/profile/view/leovineknight Looking through the Windows of madness (2009) https://www.amazon.co.uk/gp/product/1601458827/ref=dbs_a_def_rwt_bibl_vppi_i0 As with all good pub jokes, longer versions (to infuriate listeners and risk physical attack by drinking companions) spin the story out by repeatedly increasing the amount of the wager until the victim accepts it. Enjoy! John F Hall [Retired academic survey researcher] IBM-SPSS Academic Author 9900074 Email: [hidden email] Website: http://surveyresearch.weebly.com/ SPSS course: http://surveyresearch.weebly.com/1-survey-analysis-workshop-spss.html Research: http://surveyresearch.weebly.com/3-subjective-social-indicators-quality-of-life.html From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Jeff A Thanks Jon, Actually I like the extra extensions for myself, but my social-science/math-code-phobic students can’t deal with the complexity. I’m experimenting this year with illustrating a number of things using hypothetical data created with the “Make Cases” dialog you provided before and it seems to be working well so far, but that’s not for the students. Having them run their own correlation matrixes and having them run several regressions is working much better than trying to illustrate things like control variables in some other way, but I need to keep the syntax simple for them. Thanks for the assistance. Jeff From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Jon Peck Statistics tends to give the output that is most commonly wanted and then provide editing facilities to adjust further such as the pivot table editor. In this case, pivoting does the job, and that can be automated with this syntax. SPSSINC MODIFY TABLES subtype="'Correlations'" SELECT=0 /STYLES CUSTOMFUNCTION="customstylefunctions.moveRowsToLayers(fromdim=0,todim=0)". I know you don't like these extension commands and post edits, but this extension command can enhance correlation matrices in a number of ways, including blanking the upper triangle, hiding insignificant correlations, and bolding large ones. On Mon, Apr 8, 2019 at 3:18 PM Jeff A <[hidden email]> wrote:
-- Jon K Peck ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
Re: Can a correlation matrix be printed easily ?
|
Administrator
|
In reply to this post by Jeff A
Has anyone had their students try to install extension commands in university
computer labs? Does it work? It would not surprise me if student (and ordinary faculty) accounts lacked needed admin rights for such installations. Jeff-2 wrote > Thanks Jon, > > > > Actually I like the extra extensions for myself, but my > social-science/math-code-phobic students can’t deal with the complexity. > > > > I’m experimenting this year with illustrating a number of things using > hypothetical data created with the “Make Cases” dialog you provided before > and it seems to be working well so far, but that’s not for the students. > Having them run their own correlation matrixes and having them run several > regressions is working much better than trying to illustrate things like > control variables in some other way, but I need to keep the syntax simple > for them. > > > > Thanks for the assistance. > > > > Jeff > > ----- -- Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
|
|
It likely wouldn't work in our teaching labs and definitely wouldn't stay across sessions. All settings are returned to the default every time the computer is rebooted for a new user to log in. I don't see this as a disadvantage since I don't have to worry about leaving anything on the computer after I leave - it all goes away as soon as I shut the machine off. ...no problems at all with my office machine where the program is locally installed. ...but we have some pretty good IT folks so I'm sure that these types of details vary widely across universities.
Jeff -----Original Message----- From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Bruce Weaver Sent: Tuesday, April 9, 2019 9:40 PM To: [hidden email] Subject: Re: Can a correlation matrix be printed easily ? Has anyone had their students try to install extension commands in university computer labs? Does it work? It would not surprise me if student (and ordinary faculty) accounts lacked needed admin rights for such installations. Jeff-2 wrote > Thanks Jon, > > > > Actually I like the extra extensions for myself, but my > social-science/math-code-phobic students can’t deal with the complexity. > > > > I’m experimenting this year with illustrating a number of things using > hypothetical data created with the “Make Cases” dialog you provided > before and it seems to be working well so far, but that’s not for the students. > Having them run their own correlation matrixes and having them run > several regressions is working much better than trying to illustrate > things like control variables in some other way, but I need to keep > the syntax simple for them. > > > > Thanks for the assistance. > > > > Jeff > > ----- -- Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
|
|
If someone wanted some extension commands available in such a lab, it would make sense to install them as part of the original setup so they would persist through reboots. And whether admin rights are needed to install extensions depends on how the computer security is configured and where the extensions are installed. One could set the environment variables for custom dialogs and extensions in the initial configuration, which would avoid the need for admin rights except perhaps in the most extreme locked down configurations. On Tue, Apr 9, 2019 at 7:17 AM Jeff A <[hidden email]> wrote: It likely wouldn't work in our teaching labs and definitely wouldn't stay across sessions. All settings are returned to the default every time the computer is rebooted for a new user to log in. I don't see this as a disadvantage since I don't have to worry about leaving anything on the computer after I leave - it all goes away as soon as I shut the machine off. ...no problems at all with my office machine where the program is locally installed. ...but we have some pretty good IT folks so I'm sure that these types of details vary widely across universities. |
|
|
In reply to this post by Bruce Weaver
Given the other thread I'm on re: difficulty installing extensions, thought I'd chime in. My students and I are able to download extensions from the hub, and they are permanently downloaded. The University of Michigan uses AppsAnywhere which is cloudbased, so maybe that allows for more freedom than if the programs were resident on local machines.
Brian
From: SPSSX(r) Discussion <[hidden email]> on behalf of Bruce Weaver <[hidden email]>
Sent: Tuesday, April 9, 2019 7:39:41 AM To: [hidden email] Subject: Re: Can a correlation matrix be printed easily ? Has anyone had their students try to install extension commands in university
computer labs? Does it work? It would not surprise me if student (and ordinary faculty) accounts lacked needed admin rights for such installations. Jeff-2 wrote > Thanks Jon, > > > > Actually I like the extra extensions for myself, but my > social-science/math-code-phobic students can’t deal with the complexity. > > > > I’m experimenting this year with illustrating a number of things using > hypothetical data created with the “Make Cases” dialog you provided before > and it seems to be working well so far, but that’s not for the students. > Having them run their own correlation matrixes and having them run several > regressions is working much better than trying to illustrate things like > control variables in some other way, but I need to keep the syntax simple > for them. > > > > Thanks for the assistance. > > > > Jeff > > ----- -- Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
| Free forum by Nabble | Edit this page |