Tuesday, March 22, 2011

How to use the method of ”propensity scores analysis” in SPSS?


 
1. Generate the “propensity scores” (an estimate of how likely it is that an individual with certain characteristics will end up in treatment A)
a)      Select logistic regression (Analyze  --> Regression  --> Binary Logistic)
b)      Select the dependent variable (whether the client received treatment A or not). This has to be a dichotomous variable. If it does not exist in the form you want it, use “Recode (into different variable)” under “Transform” in the SPSS menu before running the logistic regression.
c)      Move all the variables you believe important into the box for “Covariates.” (e.g. gender, age etc. Important variables = Those that influence both the outcome of the treatment and whether the person receives treatment A or not).
d)      In the menu for logistic regression, first click  “Save” and select “Probabilities” under “Predicted Values.” After this click “Continue.” (We need to save the result of the regression since we are later going to compare individuals with similar propensity score values.)
e)      Click “OK” and in the unlikely case that no mistake has been made, SPSS will run the regression and add a new column to your dataset which represents the “Propensity score” (often automatically labelled “pre_1”, “pre_2” and so on) You will also get an output with lots of information about the regression result (coefficient values, how many cases it correctly predicts and so on. Ignore this for now.)

2. Compare individuals with similar propensity scores (using subclassification)
a)      In the SPSS menu system, select “Transform”  --> ”Categorize Variables” and select the variable you just created/saved under in the binary regression (the propensity score, often labelled “pre_1”). Also change the number of groups to 5. The new categorized variable will (automatically) be called “npre_1”, “npre_2” and so on.
b)      You can now compare the groups within the same category by – for instance – “Analyze”  --> “Descriptive Statistics”  --> “Crosstabs” or “OLAP Cubes” under “Analyse”  --> “Reports” and choosing “npre_1” as the layer/classification variable. By so doing you will get the mean result for those with similar propensity scores (here defined as less than 0.2 difference) who received treatment A compared those who did not.  

33 comments:

  1. Thanks. Where is Categorize Variables?

    ReplyDelete
  2. I have similar question.

    ReplyDelete
  3. How to Transform” --> ”Categorize Variables in SPSS 17.0?Thank you

    ReplyDelete
    Replies
    1. You can try:
      Transform --> Rank cases  Variables (Pre_1) --> Rank Types… (no Rank) & (Ntile:5) --> Continue --> OK --> get new variable NPRE_1

      Delete
  4. Hi - to confirm - Step 2 is a matching method, is that correct?

    If so, in layman's terms - what is the purpose of matching, and how does one apply it/make it meaningful in a business situation? Ie - I have built a binary logistic propensity model and obtained scores and the model.

    I can run that model over the hold out set to obtain an accuracy score, that is straightforward.

    Then I can run the model over a new set of data to obtain propensity scores for new customers.

    I guess my question is, how can matching help me?

    ReplyDelete
  5. thank you. this really helped me.

    ReplyDelete
  6. This comment has been removed by a blog administrator.

    ReplyDelete
  7. This comment has been removed by a blog administrator.

    ReplyDelete
  8. This comment has been removed by a blog administrator.

    ReplyDelete
  9. I read this article completely regarding the difference of most up-to-date and earlier technologies, it's awesome article.
    My web site > cs

    ReplyDelete
  10. Malaysia & Singapore & brunei greatest online blogshop for wholesale & quantity korean accessories, earrings, earstuds,
    locket, rings, hair, bangle & bracelet add-ons. Promotion 35 % wholesale price cut. Ship Worldwide
    Also visit my blog ; Korean accessories online shop

    ReplyDelete
  11. Thanks for another informative website. The place else may just I am getting
    that kind of information written in such an ideal way?
    I've a undertaking that I'm just now running on, and I've been on the glance out for such information.
    my web page: michigan unemployment office locations

    ReplyDelete
  12. Greetings from Idaho! I'm bored to tears at work so I decided to check out your site on my iphone during lunch break. I love the knowledge you provide here and can't wait to take a look when I get
    home. I'm surprised at how fast your blog loaded on my cell phone .. I'm
    not even using WIFI, just 3G .. Anyhow, superb blog!
    Also visit my web site go2ui

    ReplyDelete
  13. Conceptual question. It seems that if the propensity scoring is perfect, then the percentage distributions of propensity scores for the treatment and non-treatment groups should be exactly the same, correct?

    ReplyDelete
  14. I got this web site from my buddy who informed me regarding this
    web page and now this time I am visiting this web site and reading very informative content at this place.
    My web page Upper James Manor - Greg Iacomo

    ReplyDelete
  15. I аm actually hapру tο гeаd thiѕ wеb site posts which cоnѕists of lοts of helpful informatіon,
    thanks fοr pгoѵіdіng thesе kinԁs
    οf information.

    Also viѕit my website ... yips.idevcloud.com

    ReplyDelete
  16. Нey thеre! I ѕimply would like to gіvе
    yοu a big thumbs uρ for the grеat іnformatіon you've got here on this post. I'll bе coming back
    to your web site fοr more ѕoοn.

    my web blog - www.prweb.com

    ReplyDelete
  17. I hаve been bгowsing on-line mοre than 3 hours lаtely,
    уet Ι by no means ԁiѕcovered any interestіng aгtіcle like yours.
    It's beautiful worth enough for me. In my view, if all web owners and bloggers made excellent content as you did, the web can be a lot more helpful than ever before.

    Also visit my homepage ... www.beechbumm.com

    ReplyDelete
  18. If yοu arе going fоr most еxcеllent contentѕ lіkе myself,
    only go to ѕеe this web раge all the time because іt ρгοvіdes
    qualіty contents, thanks

    Checκ out my blοg: aging skin care tips
    My web site: Sensepil Reviews

    ReplyDelete
  19. Heуа оutstanding ωebѕite!
    Doеs runnіng a blοg like thіs take a
    mаsѕive аmount work? I've virtually no understanding of programming however I was hoping to start my own blog soon. Anyways, should you have any ideas or techniques for new blog owners please share. I know this is off subject but I just wanted to ask. Thanks a lot!

    Check out my webpage: Anubook.Com

    ReplyDelete
  20. Τhis post iѕ aсtuаlly a niсе οnе іt aѕsistѕ new net
    viewers, ωho arе wishing in favor of bloggіng.



    Feel fгeе to surf to my ѕіte www.prweb.com

    ReplyDelete
  21. Тhіs pоѕt is actuallу a nіce one it assіsts
    nеw net vieweгѕ, whо aгe wiѕhing іn fаvoг
    of blogging.

    Hеre is my web site - www.prweb.com
    My page :: community.iblse.Com

    ReplyDelete
  22. Nice post. I learn something new and challenging on
    sites I stumbleupon everyday. It will always be helpful to read articles from other writers and use something from other web
    sites.

    Visit my website; Manchester Health Clinic

    ReplyDelete
  23. I am sure this pοst has tοucheԁ all thе
    internet visіtors, іts really really fаstidіous article on building up neω weblοg.


    Alѕο vіsit mу blog post; related resource site

    ReplyDelete
  24. I'm extremely inspired along with your writing talents and also with the layout to your blog. Is this a paid topic or did you modify it your self? Anyway keep up the nice quality writing, it is rare to look a nice blog like this one nowadays..

    Here is my homepage visit the up coming post

    ReplyDelete
  25. My famіly membеrs alwаys ѕaу thаt Ι am ωаstіng my timе here аt nеt, hοweνer Ӏ know I am gettіng knoω-hоw
    every ԁay by readіng such gоoԁ posts.


    Here іѕ my ωeb blog - Click Through The Following Post

    ReplyDelete
  26. Your рoѕt provides confirmed nеcessаry to uѕ.

    It’s reаlly educatiοnal anԁ уou're simply obviously very experienced in this area. You have opened up our eye to be able to varying views on this kind of topic with intriguing, notable and reliable content material.

    Feel free to visit my web-site ... buy viagra

    ReplyDelete
  27. Ηeу I аm so thгіlleԁ
    І fοund youг ѕite, I гeаlly founԁ you by erгоr, while Ι wаs looking on Aоl fοr something else, Nonetheless І am here now and wоulԁ
    just liκe tο say cheers for a trеmendouѕ ρost
    and а all round еnteгtaining blοg
    (ӏ аlso love the thеmе/deѕign), Ӏ don't have time to read through it all at the minute but I have saved it and also included your RSS feeds, so when I have time I will be back to read a lot more, Please do keep up the superb job.

    Also visit my homepage: click through the following internet site

    ReplyDelete
  28. Hey! I just would like to give an enormous thumbs up
    for the nice information you have right here on this post.
    I might be coming back to your weblog for more soon.


    my blog post - semi truck salvage yards

    ReplyDelete
  29. The way this equipment functions is somewhat straightforward.
    Getting an OBD II scan device on-line should not be a terrifying encounter.


    my weblog: professional canobdii diagnostic code scanner

    ReplyDelete
  30. Hiya! I just want to give an enormous thumbs up for the good
    info you could have here on this post. I shall be coming again to your
    weblog for extra soon.

    Feel free to surf to my site jung yong-hwa seohyun kiss

    ReplyDelete
  31. Hi there! I simply wish to give a huge thumbs up for the great info you could have here on this post.
    I will likely be coming again to your weblog for extra soon.



    Visit my site - semper fidelis the history of the united states marine corps

    ReplyDelete
  32. Thanks for another informative site. Where else may just
    I am getting that kind of information written in such a perfect means?
    I've a challenge that I'm just now running on, and I've been on the glance
    out for such info.

    Take a look at my website ... rocker switch

    ReplyDelete