I-129F Delay at the CSC: A Comprehensive Analysis of the Data

[I & B 273]

Introduction:

So many have openly wondered what exactly has been causing the recent delay in the processing of I-129F petitions for K-1 Fiancé Visas at the USCIS’s California Service Center (“CSC”). Many on this board have done great work in pointing out that there seems the recent slowdown coincides almost perfectly with the CSC’s acceptance of a new type of application called the I-821D Consideration of Deferred Action for Childhood Arrivals (herein after “DACA” applications). But, due to the limited way in which the USCIS publishes data for its service centers through its online dashboard, it has been difficult to confirm whether or not these applications really were having this impact. In particular, the data published by the USCIS is only on a per-application type, per-service center basis, meaning it is difficult to aggregate across different application types at a particular service center. Moreover, the DACA application data information is not available through the same dashboard service and has been published separately in PDFs.

This weekend, however, in an attempt to turn my frustration into productive use, I have finally aggregated all of the relevant data for all petition types at the California Service Center and thoroughly analyzed it. The conclusion, as you will see detailed below, is incontrovertible: Yes, I-129F petitions have been directly and disproportionately delayed due to the CSC’s receipt and processing of DACA applications since August, 2012.

Methodology:

You can access the excel spreadsheet I created for all of this here, though I warn you it may be difficult to follow the method to my madness after the fact. My organization works for me, but your mileage may vary.

To conduct the following analysis, I first aggregated all of the data for all of the application types processed at the CSC. This includes I-102, I-129, I-129F, I-130, I-130 Pref, I-131, I-140, I-360, I-485, I-526, I-539, I-751, I-765, I-817, I-821, I-824, and I-90 petitions. It also includes the data for DACA applications, as well as some select data for national application trends and I-129F petitions made to the Vermont Service Center (VSC). All of the data collected was recorded monthly and had a range of 24 months, from Feb., 2011 until Jan., 2013.

I then calculated some basic statistical figures from that raw data, including means, medians, standard deviations, and variances. I also calculated two ratios for each application type for each month recorded. Those ratios are the Completions Per Pending (“CPP”) ration, which tells you how many completions occurred as a percentage of the number of applications that were currently in the backlog for a given month, and the Completions Per Receipts, which tells you how many completions there were per the number of new petitions received in a given month. I believe both numbers give valuable information about the ebb and flow of the application intake and processing processes.

Next, I focused the statistical analysis on trying to understand the differences between two periods—before and after the CSC began accepting DACA applications. The applications were first accepted in August, 2012 (though only for half the month) and remained accepted through Jan., 2013. The two periods, then, are Feb., 2011 through July, 2012 and Aug., 2012 through Jan., 2013.

With all the data laid out and statistics calculated, I then set out to create some graphs to simplify and present it in its raw form. Those are all included below in the graphs section.

Finally, I used the Gretl software suite to run a series of linear regressions on the data. I have included the most pertinent of those, along with my analysis of it, below as well.

Graphs:

1. Trend Graphs I: Big Trends

You can see here that there has been a huge fall-off in the completion of I-129F apps at the CSC since Aug., 2012. Completions remain far below the two year average and all measures of completions remain below the lowest point prior to Aug., 2012.

There was a similar national trend in completions slow downs starting at Aug., 2012, but it was resolved by November, 2012 and application completions were back above their two year average soon after.

2. Trend Graphs II: The CSC In-Depth

Meanwhile, there has been no similar fall in completions for non-DACA and non-I-129F applications at the CSC, suggesting that I-129Fs alone are bearing the brunt of the DACA application workload.

By adding the I-129F petitions into the CSC application data, we see that completion ratios are depressed, further suggesting the I-129F petitions are currently an especially low priority to the CSC.

If we take out the I-129Fs and add the DACA applications in, we see an uptick in the completion ratios, suggesting that DACA application completions are likely getting priority over the I-129Fs.

Finally, if we add the I-129F petition data back in with the DACA application data also included, we again see how the I-129F completions depress the ratios. This indicates that the I-129Fs are receiving a lower priority for completion than the other petitions.

3. CSC CPP & CPR Comparisons Over Time

Here we can see that, in fact, I-129F petitions have historically been completed at a lower rate per their backlog than other applications at the CSC, suggesting at first glance that the current depressed CPP ratio might not be such an anomaly. Looking, however, at the data since Aug., 2012 suggests there is a current divergence from the historical pattern caused by the DACA apps. (Note also that the completions per pending figure may just indicate that, compared to other apps, the CSC just operates by holding a larger backlog than for other apps normally, which may not necessarily affect the overall processing time of a given I-129F petition so long as the ratio stays relatively constant.)

What’s even more interesting, however, is that this graph shows that the DACA applications have not had any noticeable impact on the completion rates for any other application except the I-129F. Whether you include completions of DACA apps or not, non-I-129F All CSC Apps CPP ratio stays the same. Pretty damning evidence that the DACA apps are getting the same priority as every other application type and that I-129Fs are being singled out for a slowdown to compensate for that.

Here we see that, on the basis of Completions Per Receipts--which is perhaps a more useful figure for determining the relative priority given to a particular application type in a given month—I-129F petitions have historically been dealt with at roughly the same rate as other filings at the CSC. But, again, there is a major divergence since Aug., 2012 that suggests I-129F petitions suddenly got pushed to the back burner while other app types, including the newly filed DACA applications, got increased attention.

[Continued in the next post due to limit on image embedding per post...]

[I & B 273]

[Continued from the previous post due to limit on image embedding per post...]

4. Effects of Including the DACA Apps in the CSC Total Workload Stats

So let’s look more closely at what happens when we include the DACA application data to the other CSC data. The number of pending applications rises, the number of overall completions rises as the office starts turning out DACA approvals, the increase in the number of receipts is of course evident, the CPP decreases briefly before re-converging (note from the above graph how we know this decrease comes on the back of 129F and the subsequent convergence is improvements in DACA completions, not a fixing of the 129F completion figure), and the CPR does the same as the CPP (with the same underlying mechanism at work).

All Others Pending, Completions, and Receipts trends from above, but with percentage increase figures also included for perspective.

CPP and CPR from the first effects graph, but with percentage change figures added for perspective.

5. Comparisons of Pre-Aug., 2012 Stats with Post-Aug., 2012 Stats

We can see here that the slowdown in I-129F processing times isn’t being driven by an abnormal surge in the amount of I-129F petitions pending. In fact, there was, on average, fewer pending there than at times prior to the DACA applications. Moreover, all other apps at the CSC are similar down in terms of the number pending. But, when you add the DACA apps in, you can see how they are swamping the CSC, raising the total apps pending to 19% higher than its pre-DACA average.

Here we see the true weight of CSC processing choices on the I-129F completions. Compared to prior to the DACA apps, the average number of I-129F petitions completed at the CSC is down a staggering 33%, a change that stands-out from the other application types, and in stark contrast to how I-129Fs are being completed at the VSC.

This graph emphasizes that the slowdown in I-129F processing is not caused by some general surge in application receipts. To the contrary, even when you included the DACA applications, overall receipts are down across the board compared to the pre-DACA period. This is pretty damning for the USCIS’s handling of I-129Fs following the receipt of DACA applications. They have handled more filings in the past.

Again, we can see that the I-129F petitions are bearing a disproportionate reduction in completions per pending apps when compared to other application types, though all application types at the CSC have taken a bit of a hit. This is measuring how big the building backlog has gotten. No such slowdown at the VSC, however.

Again, we can see that the I-129F petitions s are bearing a disproportionate reduction in completions per pending apps when compared to other application types. This is how a backlog is born and perpetuated. No such slowdowns for other application types at the CSC, VSC, or even nationally. Do you need more evidence that I-129F petitions are being singled out for this current slowdown?

Linear Regression Analysis to Determine Statistical Significance of the Correlation

Going beyond what intuition and commonsense should tell us about the above data, we can find more evidence for the hypothesis that the DACA applications have causes a statistically significant slowdown in I-129F petition processing by breaking out the big econometric guns and turning to Gretl.

Given the time series nature of the data, I chose to model a linear AR1 regression to capture the interplay between the variables overtime. What this means in layman’s terms is that I measured the relationships between key variables, both within their current period and with a one month lag effect, to see which variables were correlating with which other variables. In particular, I wanted to ascertain the extent to which other variables were effecting the CSC’s I-129F completion numbers throughout the 24 month period. There are some limitations to making this type of equation with the data I have because related variables (such as receipts and all other pending) cause co-correlation problems; however, I do believe the results I was able to generate by carefully choosing key variables are pretty damn convincing.

What this basically says is that the numbers of DACA applications pending and completed have a statistically significant negative correlation with the completion of I-129F petitions at the CSC. Correlation is not causation, but, I think given the other analysis I provided above, we can pretty much call this one. Moreover, the model used ended up being surprisingly robust, explaining 88% of the variation in the dependent variable (completions of I-129F petitions at the CSC). For your reference, here is a mapping of the regression versus the actual trend line. The residual, a graph of which I cannot include due to the images per post requirements, decreases on average post-Aug., 2012, further giving validity to the notion that the regression is correctly capturing the influence of the DACA applications. I rest my case.

Conclusion

The DACA applications have caused a major slowdown in the processing of I-129F petitions at the CSC. The CSC has singled out I-129F petitioners to bear the brunt of these DACA filings. But it doesn't have to be that way. Let your representatives know that you do not deserve to be singled out for a delay in reunited with your loved one while DACA applicants jump the line to approval. Use this evidence in your complaints to them if you'd like. The USCIS is playing games with our petitions, but now we have the facts to call them out on it. So what are you waiting for?

Edited March 25, 2013 by I & B

[cazzers 83]

This thread is really interesting and really thorough. Thank you for the hard work!

[I & B 273]

This thread is really interesting and really thorough. Thank you for the hard work!

You're very welcome. Now, if only someone who could affect the process would pay attention to these figures. It really couldn't be clearer that I-129Fs alone have been slowed down so that DACAs may be processed more quickly. Please feel free to pass it on to your congressman or senator.

[cazzers 83]

My fiance will try to go to the Congresswoman's office in the next weeks and i'm gonna tell him to print this and give it to them. I am really hoping it will help:) I just bookmarked this thread so that i don't lose it

[Bayareaguy 156]

Nice work!

I'm not very adept at statistical analysis, so I avoided doing any of that. And yeah, it's nice that it confirms that other application types are not being affected like the I-129F. It makes you wonder if there was a particular conscious reason behind the decision, or if it just because they have always had the lowest priority.

[cazzers 83]

Nice work!

I'm not very adept at statistical analysis, so I avoided doing any of that. And yeah, it's nice that it confirms that other application types are not being affected like the I-129F. It makes you wonder if there was a particular conscious reason behind the decision, or if it just because they have always had the lowest priority.

My opinion is this...when they were approving I-129f's in 3 months or less in july/august/sept, they knew about the DACA applications and that they would get a lot of them..now, i'm guessing, they didnt knew how much CSC would receive and that it would create such a big backlog. I'm still hoping that by moving DACA files to TSC is gonna get the I-129f application back on track, but from what i've seen in the last 2 weeks, my hopes are getting smaller and smaller.

[I & B 273]

My opinion is this...when they were approving I-129f's in 3 months or less in july/august/sept, they knew about the DACA applications and that they would get a lot of them..now, i'm guessing, they didnt knew how much CSC would receive and that it would create such a big backlog. I'm still hoping that by moving DACA files to TSC is gonna get the I-129f application back on track, but from what i've seen in the last 2 weeks, my hopes are getting smaller and smaller.

I think a really nuanced reading of all the data would be that the CSC was actually somewhat caught off guard by the DACA apps. You can see that there an impact on their completion rates in August, September, and October across all application types. But then they figured out an easy trick: just switch all the I-129f staff over to DACA duty, leave everything else the same, and worry about I-129Fs later. From their perspective, this cheap trick makes sense: instead of getting complaints from every type of applicant, now they could contain those complaints to just 10,000 pending apps. But the problem is that "later" never came. They never got out from under the DACA apps and so they never got back to the I-129Fs. And they didnt have a good back up plan, so now I-129F petitioners are screwed unless the purported DACA app transfer to TX really was substantial.

[cazzers 83]

I'm wondering if they transferred all DACA applications or just a part of them. Their website says "On March 8, 2013, USCIS transferred some casework from the California Service Center to the Texas Service Center to balance overall workload with processing capacity at the centers. The affected casework includes the following forms:

I-821D, Consideration of Deferred Action for Childhood Arrivals

I-765, Application for Employment Authorization (when filed concurrently with the I-821D)

I-765 Worksheet (PDF) (when filed concurrently with the I-821D)"

I'm hoping all of them and I just hate when they are not saying things exactly how they are. Or maybe I don't understand cause English is my second language?

In the end, all this analyzing about these delays won't matter much until the CSC director says: "guys, lets get back on those hunny bunny petitions"

Edited March 25, 2013 by cazzers

[Andreea&Kevin 38]

How did you account for DACA numbers not being broken down by service center? I have not seen any exact numbers of receipts and completions by service center, only the national total.

[Bayareaguy 156]

How did you account for DACA numbers not being broken down by service center? I have not seen any exact numbers of receipts and completions by service center, only the national total.

That is one thing that bothers me about the whole situation...why is the data not presented the same way? Its like it is intentional so it makes analysis more difficult. Are they being opaque on purpose?

[Andreea&Kevin 38]

That is one thing that bothers me about the whole situation...why is the data not presented the same way? Its like it is intentional so it makes analysis more difficult. Are they being opaque on purpose?

That's why I never tried to go this far in analyzing the data, just too many unknowns in the DACA data. It's obvious that CSC got hit with more DACA than any other service center with how many filers are from California. But, did they get 40% of all apps, did they get 75% or some other amount? We also have no idea where they are at with the apps they got. Now with the transfer of "workload" to TSC we know even less.

I'm very lucky to have escaped the black hole of CSC and wish the best of luck to everyone still dealing with it.

[Bayareaguy 156]

I'm very lucky to have escaped the black hole of CSC and wish the best of luck to everyone still dealing with it.

I'm glad to see you aren't criticizing us for being impatient now that you safely have your NOA2, like some here (and some with some very short NOA1->NOA2 times)

[cazzers 83]

I'm glad to see you aren't criticizing us for being impatient now that you safely have your NOA2, like some here (and some with some very short NOA1->NOA2 times)

[Azsara 110]

very nice work, thank you !