Error in reclink

[Tarandeep Kaur]

I am using reclink to fuzzy match village names in Andhra Pradesh, India.
Code: use "C:\Groundwater\matching village names MI6 and JJM -- Andhra Pradesh\unmatchedMI6.dta", clear
Reclink Blockname Villagename using "C:\Groundwater\matching village names MI6 and JJM -- Andhra Pradesh\unmatchedJJM.dta", idmaster(id_master) idusing(id_using) gen(myscore)
Result: 3420 perfect matches found
Going through 8860 observation to assess fuzzy matches, each .=5% complete ...........) required r(100);

Please tell me what error I am doing.
Thanks in advance!

[Dirk Enzmann]

Welcome to the Stata Forum. Before posting you are advised to read the Forum's FAQ -- ideally fully, but most relevant here: #12. There you could learn that you should

1. mention that the command (program) -reclink- you are using is from SSC (if it is)
2. show an example of your data (via -dataex-, for more details or "how to" see the FAQ #12.2)
3. use code delimiters for code, results, and data (additionally see here)

I would suggest that you reply to your own topic to improve your question using this advice.

I am not familiar with -reclink- (obviously from SSC). However, the error code r(100) might give you a hint (although it might be a problem of -reclink- itself, which will not help much but could be investigated more closely with -set trace on-, see -help trace-). I believe that you can always click on the error code shown in the results window in blue with further explanations. Otherwise, you can use -help error-, go to the PDF-documentation and look up the meaning of the error code 100 by yourself.

[Clyde Schechter]

Unfortunately, error code 100 is "invalid syntax," which is too broad to be very helpful. In addition, the problem is likely to reflect a bug in -reclink- itself, because if the -reclink- command you issued contained a syntax error, you would not have gotten any output from the command at all (other than the error message). You can -set trace on- before the -reclink- command and run that to see where inside -reclink- the error is coming from. You may or may not then be able to figure out a fix. It may turn out to be some incompatibility with recent versions of Stata (-reclink- is very old: see next paragraph).

I have a different suggestion. -reclink- was written a long time ago and has not been updated since at least 2010. The author, Michael Blasnik, is no longer active on Statalist, so you may or may not be able to reach him, and the odds are good that even if you do, he will not be interested in fixing the program now. May I suggest that you install and use, instead, Julio Raffo's -matchit- program. It is available from SSC, and it was most recently updated in 2020, so it is likely to play nicely with recent versions of Stata. I have used it with some frequency myself and found it to be flexible, fast, and trouble-free.