NIW Citation Strategy: Using Prong 2 Evidence Right

Most petitions file citations under the wrong prong

Here is a pattern that shows up repeatedly in denied AAO decisions. The petitioner submits a strong citation record, AAO accepts that the work has been read, and then denies anyway — because the citations were argued as Prong 1 evidence (“the work is important”) when the petition needed them doing Prong 2 work (“the field has invested in this person”).

Citations rarely win Prong 1. National importance is about the scope of the endeavor, not how many people cited a paper. Where citations earn their keep is Prong 2: whether you are well positioned to advance the proposed endeavor. File them there, and frame them for that job.

What citation evidence proves under Prong 2

Dhanasar’s Prong 2 asks whether the petitioner is “well positioned to advance the proposed endeavor.” Citation evidence speaks to that when it shows independent uptake, not just a number:

• Citations from groups working in the same area, with context — review articles, methods built on yours, replications
• A citation trajectory across several years, which signals sustained interest rather than one paper that briefly spiked
• The geographic spread of citing authors, which is the cleanest way to argue international reach (see our citation map tool)
• Direct adoption: code or model downloads, dataset use by other labs, patents licensed by industry or an agency
• Invited talks at venues you have not published at, especially invitations from groups you have never worked with
• Competitive grants — a funded proposal is the field putting money behind the judgment that you can deliver

Three citation mistakes AAO discounts

Counting your own citations

A raw citation total that quietly includes self-citations and citations from co-authors is not the same as independent recognition, and AAO knows the difference. Report the independent number. Google Scholar can exclude self-citations; Web of Science and Scopus separate self and co-author citations by default. Use the cleaner figure and say which figure you used.

A number with no context

“My paper has been cited many times” is treated as a number. “My paper has been cited by review articles in the field’s leading journal and by independent labs that built methods on top of it” is treated as evidence the field absorbed the work. The second framing is what Prong 2 rewards.

Do this concretely: for your top three to five papers, pull the “Cited by” list and annotate it — which entries are review articles, which are downstream methods, which come from labs with no connection to you. Attach the annotated list as an exhibit instead of a bare count.

Vague claims of global reach

“My work has been cited by researchers worldwide” with no country breakdown is exactly the kind of unsupported assertion AAO calls conclusory. International impact has to be shown, not stated — named countries, named institutions, named groups. A specific list of citing institutions is concrete in a way that “worldwide” never is.

Citation counts only mean something against field norms

There is no regulatory citation threshold under Dhanasar, and there is no honest universal number. A 30-citation paper can be major in pure mathematics and minor in machine learning. We are not going to publish field-by-field thresholds, because we do not have verified data behind them and a made-up benchmark would do more harm than good.

The practical move is to make the comparison explicit yourself. If your count is strong relative to your sub-field, say so and back it — a senior recommender stating that the figure is high for the area, a journal acceptance rate, the typical citation range for comparable work. Give the officer the yardstick. Do not leave a number floating with no context for how to read it.

The citation-map argument

If your work is cited internationally, a geographic citation map is the cleanest single way to present that under Prong 2. It converts “cited worldwide” into a specific, checkable exhibit: which countries, which institutions, which groups.

Our free citation map tool takes a Google Scholar ID and produces a downloadable world map of citing institutions. It is specific enough to function as real evidence rather than the boilerplate an officer skims past.

How to present citation evidence in the petition

A structure that keeps the evidence doing Prong 2 work:

1. Aggregate, with the caveat. “Per Google Scholar, my work has been cited [N] times by independent researchers, excluding self-citations and co-author citations.” State the source and the exclusion.
2. Per-paper detail for the top three to five. Citation count, venue, acceptance rate or impact factor, and a sample of citing-paper titles.
3. Geographic spread. Top countries and institutions, with a citation-map exhibit.
4. Context. A few sentences on how the work has actually been used — review-article inclusion, methods built on it, datasets adopted.

Recency: old citations are fine, sustained ones are better

Age is not the problem. A paper still cited years after publication is showing exactly the durable interest Prong 2 wants. A burst of citations that then goes silent is weaker, and if your profile looks like that, the petition narrative should explain why the field still engages with the work — a follow-on line of research, a method that became standard, continued downstream use.

The one thing to do today

Pull a citation export — Web of Science or Scopus if you have access, Google Scholar with self-citations excluded if you do not — and segment it: self-citations, co-author citations, genuinely independent citations. Then audit your top three to five papers for context. That audit is the difference between a number and an argument.

Our $10 case review reads your citation evidence against the recurring patterns in approved AAO cases for your profession, and the citation map tool is free and produces a downloadable exhibit you can attach to the petition.