Timeline for Preparing survey data for analysis
Current License: CC BY-SA 3.0
19 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 22, 2015 at 10:36 | history | edited | Ayalew A. | CC BY-SA 3.0 |
added 22 characters in body
|
| Sep 8, 2014 at 5:24 | history | edited | Ayalew A. | CC BY-SA 3.0 |
deleted 2 characters in body
|
| Aug 28, 2014 at 8:33 | history | edited | Ayalew A. | CC BY-SA 3.0 |
added 8 characters in body
|
| Aug 22, 2014 at 12:12 | history | edited | Ayalew A. | CC BY-SA 3.0 |
added 1 character in body
|
| Aug 22, 2014 at 10:34 | comment | added | Ayalew A. | @StasK: Yes, now with the help of the commands you suggested I have been able to figure out and correct the coding problems that resulted in duplicate single-unit strata. Regarding the missing weight values, they resulted from missing design information for some study participants. I am considering multiple imputation to deal with the item-missing data in my dataset. Some of the weights as you have seen are less than one because I used a normalized version of the analysis weight. Thank you so much. | |
| Aug 21, 2014 at 21:26 | history | edited | StasK | CC BY-SA 3.0 |
code formatting
|
| Aug 21, 2014 at 21:25 | comment | added | StasK | At any rate, I think you have some improperly coded units, and that you will have to figure out with your sample data provider or the field staff. | |
| Aug 21, 2014 at 21:24 | comment | added | StasK |
Did it help you to figure out what the issues were? The weight obvously should never be missing. (And, if done right, weights are inverse probabilities of selection, should be no less than one, and sum up to the population size.) The missing design variables in the first four observations probably mean that you don't have any survey observations, although you'd have to check. May be I should have suggested sort PSUSTRATA PSEUDOPSU PSEUDOSSUST SSU NWEIGHT ID before listing, so that it's easier to interpret.
|
|
| Aug 21, 2014 at 15:32 | comment | added | Ayalew A. | @StasK, now I have included in the original post the output I obtained after issuing the series of commands you suggested. | |
| Aug 21, 2014 at 15:27 | history | edited | Ayalew A. | CC BY-SA 3.0 |
added 476 characters in body
|
| Aug 21, 2014 at 14:51 | comment | added | StasK | @RobertoFerrer, it's on topic per whuber: meta.stats.stackexchange.com/a/2150/5739 | |
| Aug 21, 2014 at 14:48 | comment | added | StasK | The problems are obviously in your code that creates the survey variables from whatever you obtained from the field. You'd have to review and resolve the offending units that must have been mislabeled. | |
| Aug 21, 2014 at 14:45 | comment | added | StasK |
egen _svy_tag = tag( PSEUDOPSU PSUSTRATA SSU PSEUDOSSUST FPC1 FPC2), missing [ENTER] replace _svy_tag = _svy_tag | missing(NWEIGHT) [ENTER] list PSEUDOPSU PSUSTRATA SSU PSEUDOSSUST FPC1 FPC2 NWEIGHT if _svy_tag [ENTER] and add the output to your original post. (+1 to Roberto: this is a live website where you can edit a lot of stuff -- you can certainly edit your own posts).
|
|
| Aug 21, 2014 at 13:32 | history | edited | Ayalew A. | CC BY-SA 3.0 |
deleted 8 characters in body
|
| Aug 21, 2014 at 11:53 | comment | added | Roberto Ferrer | Do not use the comment section to give additional information that is not suited for this section. Use the edit button to edit your original post. | |
| Aug 21, 2014 at 6:18 | comment | added | Ayalew A. | This is the command I issued: svyset PSEUDOPSU [pweight=NWEIGHT], strata( PSUSTRATA) fpc(FPC1) || SSU, strata( PSEUDOSSUST) fpc(FPC2). Subsequently, I issued the command 'svydescribe, final'. [I cannot include the output due to limitation of number of characters for comments on this site.] In the output I see several duplicate strata with single units where as in the actual dataset there are no strata with single units. | |
| Aug 21, 2014 at 3:19 | answer | added | Anthony Damico | timeline score: 1 | |
| Aug 20, 2014 at 16:41 | comment | added | Roberto Ferrer | I'm not sure this is on-topic, but anyway, you should include the exact commands you issued and exact output. A complete, reproducible example, is even better. | |
| Aug 20, 2014 at 16:12 | history | asked | Ayalew A. | CC BY-SA 3.0 |