If you anticipate being a victim, then a victim you will be.

I have seen this experiment recounted by a number of people in various interviews over the past year or so.  Here is Konstantin Kisin:

Double click to enlarge.

Sounds plausible but I have never seen the source study.  El Gato Malo provides it in DEI/CRT do not stop racism, they cause it.  The subheading is a look at the psychological underpinnings of grievance cults.   

The original study is Perceptions of the impact of negatively valued characteristics on social interaction by Robert Kleck and Angelo Strenta.  It is from 1980.  From the Abstract:

24 female undergraduates were led to believe that they were perceived as physically deviant in the eyes of an interactant when in fact they were not. Following a brief discussion, they commented on those aspects of the interactant's behavior that appeared to be linked to the deviance. Ss who thought that they possessed negatively valued physical characteristics found strong reactivity to the deviance in the behavior of their interactant, whereas those with a more neutrally valued characteristic did not. An expectancy/perceptual bias explanation is advanced to account for these results, although experimental demand is also a plausible interpretation. Study 2, with 50 male and female Ss, reaffirmed that both the expectancy and the demand explanations were plausible. Study 3 with 30 female Ss used a new set of instructions devised to test the competing explanations. Results strongly undermine a demand interpretation of the original results. In Study 4, with 32 female Ss, persons who had observed the behavior of the interactants in Study 1 via videotape also perceived greater reactivity to an imputed negative form of deviance than to a neutral one. Data support the notion that the results of Studies 1 and 3 reflect the operation of an expectancy/perceptual bias mechanism and tend to rule out a self-fulfilling prophecy dynamic. (17 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)

A pdf version is here.  The Discussion section substantially mirrors Kisin's summary.  The opening of the discussion section:

The general pattern of results is consistent with our hypothesis that a negatively valued physical characteristic is perceived by its possessor as having a greater impact on the behavior of an interactant than one that is not negatively connoted. Further, there is some evidence suggesting that the type of physically stigmatizing condition involved will affect the aspects of the other individual's behavior that are scrutinized for evidence that the physical deviance is being responded to.  In the present case, for example, individuals who thought that they had a scar were more likely to focus on the gaze behavior of their interactants, whereas those who thought that the other attributed epilepsy to them were sensitive to behaviors indicating tenseness and anxiety.

So Kisin is citing a real study and citing it accurately.  But it is 44 years old and the sample sizes are small, though the methodology is reasonably strong.  

A further study, done in 1986 is Gender and responses to disfigurement in self and others. by Robert E. Kleck and Christopher A. Strenta.  

I would still be reluctant to take all this at face value.  Plausible, but not yet definitive.  I search on expectancy/perceptual bias mechanism to see if there is any further body of research and or replication of this particular experiment.  

There is a Wikipedia entry which is a good jumping off spot.  Expectancy/perceptual bias mechanism is apparently a well known, much discussed and reasonably well studied issue.  

APA Psycnet provides a study which sort of confirms the Kleck study but with a twist.  From Teacher expectations: Self-fulfilling prophecies, perceptual biases, and accuracy. by Jussim, L. (1989). 

Teacher expectations: Self-fulfilling prophecies, perceptual biases, and accuracy. Journal of Personality and Social Psychology, 57(3), 469–480. https://doi.org/10.1037/0022-3514.57.3.469.  From the Abstract:

Students' performance may confirm teachers' expectations because teacher expectations create self-fulfilling prophecies, create perceptual biases, or accurately predict, without influencing, student performance. Longitudinal data obtained from 27 teachers and 429 students in 6th-grade math classes assessed the extent of self-fulfilling prophecies, perceptual biases, and accuracy. Results revealed modest self-fulfilling-prophecy effects on student achievement and motivation, modest biasing effects on the grades teachers assigned students, and that teacher expectations predicted student performance more because they were accurate than because they caused student performance. Results provide more support for perspectives emphasizing limitations on expectancy effects than for perspectives emphasizing the power of expectancies to create social reality. They also provide more evidence of accuracy in social perception than of error and bias. (APA PsycInfo Database Record (c) 2016 APA, all rights reserved)

It confirms that there is modest self-fulfilling-prophecy effects, however, the data also provides "more evidence of accuracy in social perception than of error and bias."  Observers may form and impose expectations but this can be complicated when the observer assessments are actually accurate.  

An overall summary of the Observer Expectancy Effect is here.  

There is a related study, much more recent and more pertinent in the age of Zoom communications but dealing with a different problem.  To Look or Not to Look: Acknowledging Facial Stigmas in the Interview to Reduce Discrimination by Juan M. Madera and Mikki Hebl.  

I see in a couple of discussion forums comments to the effect that the Dartmouth Scar Experiment is well known and has been replicated a number of times.  I have not seen the replications, only the assertion that is has been replicated.  

After thirty minutes of digging around my conclusion is that Kisin offered a reasonable summary of what is apparently a well known and well regarded study which has perhaps been replicated.  Other, closely related, studies have arrived at the same or very similar conclusions as the original.

Does the expectations that one carries into an interaction create the conditions by which we will interpret the experience of that interaction?  Apparently yes and on reasonably robust and replicated grounds.  

Is Kisin on solid ground to assert that a victimhood mindset (the predicate anticipate that one will be discriminated against) increase one's self-perceived experience of discrimination?  Seems so.  

All this is closely related to Miles' Law.

"Where you stand depends on where you sit."

How you interpret things depends on your context.  If you anticipate being a victim, then a victim you will be.