The "anchor citation" is a name we use for the paper that has the scoring instructions or normative data that we use in our own scoring. This is the citation that we would include in the Method section of a poster or paper.
Often a measure will have more than one publication, and in some cases there could be thousands of papers. How do we decide which one is the anchor? It should be the one that reports the details that we followed for our project: If someone pulls the anchor citation, the items should match the questionnaire that we used exactly. Our scoring procedures should duplicate what the anchor citation describes, or we should be sure to report any alterations and the rationale for them. If the measure has gone through multiple versions of norms, then it would be best practice to report the citation that built the norms against which we are calibrating our own research study scores.
For example, the Achenbach Child Behavior Checklist has gone through multiple versions, with the 1983, 1991, 2001, and later versions each changing a few of the items and gathering an entirely new sample as the basis for the age and sex norms. Good scientific practice would involve being sure to describe which items and which norms were used. We can accomplish this by confirming that the version of the technical manual that we cite matches the version that was used in data collection and scoring.