Subscribe Bookmark RSS Feed

Need help on data

chamuh1

Community Trekker

Joined:

May 5, 2016

Hello,

I just started using JMP 12 and am having a really difficult time in using it. I ma in NYC for a few days and would be happy if someone could show me how to use its features for a data I have from a completed study. Once I have the data analyzed then I am able to write an original research article for journal publication. If anyone is free for an afternoon or whatnot then please let me know.

Thanks

7 REPLIES
txnelson

Super User

Joined:

Jun 22, 2012

I have to be honest, your request makes me very suspicious of what you are asking for.  So if you could provide more details about what the specifics of your project are, it might make it easier for me, or others to meet your needs.

.

Let's start with the basics:

  What interest area are the data from(medical, engineering, genealogy, etc.)?

  What analysis are you needing to do?  Or what question are you attempting to ask about the data.

  What form are your data currently in?

  Is it possible for you to provide a sample of your data as an attachment?

Also, there are many online self help tools available for the new user.  Also, many tutorials are provided under the Help menu in JMP.  

 

Jim
chamuh1

Community Trekker

Joined:

May 5, 2016

Hello,

The subject area is medical. A cross-sectional study in a country was done using a survey instrument. There was a total of 49 responses recorded given both online (25 responses) and on paper (24 responses). The online instrument was delivered and responses collected using surveymonkey. The data exported from surveymonkey comes in a SAS file. The written responses were entered into survey monkey as a separate file so that data could be exported as a SAS file.

The study is about sexual risk behavior and HIV infection. I want to compare the written responses versus the online responses and determine the differences between the groups, as well as characteristics in either groups (online versus in-person responses) which I assume is descriptive statistics. I am unsure what other tests would be good to show any significance.

Each time I use the SAS file in JMP I get numerous tabs and I find it confusing to consolidate the results and summary SAS files into an appropriate data file so that it is manageable.

I am new to JMP and have no idea why my questions would be suspicious to you or any other user . I figured posting on this forum would be appropriate for a novice user and that I would be pointed into the right direction. I just want to statistically understand my data, be more comfortable in using JMP, and finish writing my original article for a public health journal.

Thanks.

txnelson

Super User

Joined:

Jun 22, 2012

Thank you for your detailed response.

Let me first dispense with my comment about being "suspicious".  It is not uncommon to get individuals to post to the forum who are only looking for someone to do their work for them.  The way your question was stated, raised that red flag.  With your detailed response, it is far clearer what you are asking of the forum contributors. 

Would it be possible to share the SAS file?  It is difficult to understand what exactly would have to be done to put the files together in JMP, given the broad number of things that could be coming from SAS.  That would be the starting point.  Also, do you have a conceptual idea of how you want to match up the written responses with the online responses?  A formal platform for doing Textual mining is not available in JMP 12.  But other methods could be configured.

Jim
KarenC

Super User

Joined:

Feb 10, 2013

I am not sure about your loading the file into JMP issues. Are you sure it is a SAS file and not an SPSS file? It shouldn't matter, either way JMP should be able to read it. The SPSS files that one gets from survey tools often have addition information that gets put into the appropriate column info settings. If you have multiple JMP tables you may need to "Join" them (multiple tables with information for the same subjects where you will need to join based on the subject ID) or you may need to "concatenate" them (if say you have 2 tables, one for the online and one for the paper subjects where all of the columns are the same).  Look those terms up in the help.

Once you get your data then you will probably want to start with "distributions" and "fit y by x" with your group (online vs. paper) as your x and the other items of interest as you y's.  You may also want to look into the consumer research platform (google consumer research JMP) to find the help pages. 

Karen

chamuh1

Community Trekker

Joined:

May 5, 2016

You know, now that you mention it, it is an SPSS file. I am not sure why I thought it was a SAS file. Either way, I do get multiple table and it is overwhelming. I tried joining the tables but seemed to have gotten further lost in the process.

I will not post the original data file on this forum because there are IP address identifiers for each respondent. I understand your concerns about having someone else doing the analysis, but I really want to learn how to do it on my own (obviously with some guidance).

Karen, do you think if I make an excel file with all the appropriate data points for each table, then copy/paste them into a JMP data table would make the overall task easier? I know my suggestion might appear laughable, but I am quite lost as what exactly I should be doing to analyze the raw data. I am new to all this.

As for comparing the written responses versus the online responses, the goal is to determine the population demographics that responded in each group and the potential expansion of service by the local NGO that collaborated on this project. Not everyone on the island had internet access, and there is only one field office providing HIV/AIDS education and advocacy services.

txnelson

Super User

Joined:

Jun 22, 2012

Cutting and pasting into Excel, might be the quickest way to get to your results, given that you can not share the tables that your SPSS files are creating.   If the Excel file you create is large, it might be easier to save the Excel file, and then just open the Excel file from JMP.

I might also be able to help you put the files together if you could provide a description of the files that are being created.  Don't include the actual content, but provide the columns names created, and a description of the data in the file.

Jim
KarenC

Super User

Joined:

Feb 10, 2013

As Jim said, using excel to organize your data may work best for you at this point.  You can also cut and paste within JMP if that helps (sometimes I do find it easier simply to cut and paste items between tables rather than using the table features - depends on how much data).  With only the small number of subjects you might take one of the tables as your base table and then add to it.  You can add rows by double clicking at the bottom to the table, same for columns.  


Once you get it organize then the Fit Y by X platform is going to be a good place to start. You can also use the distribution platform and include the distribution of paper/internet.  Then if you click on one it will highlight those people in your distributions (fun way to look to see how things change by respondent type).

Good luck