Stereotype content across ethnicity and socio-economic status: a case of ethnic groups in Russia

There are, however, some differences that occur most likely due to the added influence of the status variable. Russians have second to best scores on both competence subscales - assertiveness and skills - falling behind by Chechens and Chinese, respectively.

Even though Fiske (2018) suggests that labels and photographs evoke similar stereotypes, that might not turn out to be exactly accurate for the current study as targets are sometimes mislabelled by participants, which might be the reason for more muddied up stereotype content thus leading to results of the current study not replicating results of previous studies perfectly.

H2a: High-status targets should be evaluated as more competent than warm.

Overall, high status does evoke stereotypes of competence, with high-SES targets being assessed as more competent than warm (see Table 9) as said by Koenig and Eagly (2019), but when we view the results for each target ethnicity individually (Table 11), we can see a different picture.

The general pattern is only applicable to the Chechen target. High-SES Russians areevaluatedas equally competent and warm, which does follow typical warmth-competence distribution for Russian participants as Russians are one of three groups that were viewed as unambivalently “good” on both SCM dimensions (Grigoryev, et al., 2019).Uzbeks and Chinese are viewed as more warm than competent which is novel results that are hard to explain right away. One of the possible explanations might be that, according to commentary of some of the participants some photographs, though constructed in a similar way, were subjectively more nice-looking and some of the targets (namely, Chinese) were said to be more “pretty”.

H2b: Low-status targets should be evaluated as more warm than competent.

Once again, the warmth competence trade-off seems to be evident for the overall results (Table 8), supporting the trend seen in multiple other research done in other countries (Durante & Fiske, 2017; Fiske et al., 2016; Fiske, 2017, 2018). In that case the general pattern applies to all groups other than Chechens - low-SES Russians, Uzbeks and Chinese are indeed evaluated as more warm than competent. Chechens are viewed as less warm than competent, which while not following the warmth-competence trade-off, does follow the results acquired for Chechen ethnic group earlier (Grigoryev, et al., 2019).

As for the crossed categorization predictions, I find that it will be better to discuss the following two hypotheses together.

H3a: The consistent information about stereotype content from the status and ethnicity should lead to no changes in the evaluation of targets.

H3b: The inconsistent information about stereotype content from the status and ethnicity should lead to the additive pattern with changes in the evaluation of targets toward neutral.

These hypotheses have been mostly met, with some exceptions that will be described in detail below.Similarly to previous research (Grigoryan, 2020) the additive model seemsto fitthe data showing cross-categorization results, which further supports its' legitimacy for studying stereotype content formulation.

The content of stereotypes specific to any one ethnicity is a bit different from what have been previously found for Russia. That is most likely due to combination of factors such as adding a new independent variable that, expectedly, have affected the stereotype content and possibly the type of stimuli and the sample acquired.

If we take Chechenspreviously low on competence(LC) and warmth (LW) and look at the competence and warmth dimensions for high status (HC-LW) and low status (LC-HW) then indeed, we can see that the low-SES group is evaluated as low competence (so that score does not change with consistent status and ethnicity information) and middle warmth (LW (ethnicity) +HW (status)); the high-SES group is evaluated as moderately competent (LC+HC) and cold (LW and LW (`and' between variables highlights consistent information)).

If we apply the same analysis to Russians (HC-HW), we can se that low-SES Russians are evaluated as moderate on competence scale (HC+LC) and high in warmth dimension (HW and HW); the high-SES group is evaluated as competent (HC and HC) and warm, which the first place where a prediction is not entirely accurate, as high status Russians are still high on the warmth scale. That could be arguably due to the fact that Russians tend to be rated exceptionally high in warmth when studying Russian population (Grigoryev, et al., 2019), so even though the rating is above moderate on the warmth scale here it still much lower for high status group, than for low-status groups or in a study without a status indication.

For Uzbeks (LC-LW) it is important to mention that though they are part of the cluster of ethnicities that is low on both dimensions, within this cluster they are quite close to its' edges on the moderate side. The low-SESgroup is evaluated as low in competence (LC and LC) and moderate in warmth (LW+HW); and the high-SES group is evaluated as moderate in competence (LC+HC) and moderate in warmth which is does not align with the hypothesis perfectly as the results should have been low in warmth, but might be due to its' borderline position on the SCM map.

And now we come to the last group, Chinese (HC-LW). Low-SES Chinese are evaluated as incompetent, which is unexpectedly low, and moderately warm (LW+HW); high-SES group is evaluated high both on competence (HC and HC) and warmth scales, which very surprising as a low rating was expected. It seems to be an interesting to topic to study stereotype content relevant specifically to the Chinese in a separate study, as this ethnicity seems to provide the most unexpected results when it comes to cross-categorization. A possible explanation might also hide, as have been mentioned when discussing ethnicities, in the pictures used as stimuli and their attractiveness. Some of the unexpected data described above might be a result of the fact that Russia is a post-communist country, and stereotype content of high-status groups and lower class groups in post-communist countries was found to be different from the results obtained in other countries - the lower class was precepted more positively, and the upper class - more negatively (Grigoryen et al., 2019)

Now, after discussing the hypotheses proposed, I want to highlight some other findings, that seem noteworthy to me.

Most of the studies researching the effect of status on the stereotype contents work with perception oneither group or individual level. When it comes to ethnicity (Grigoryev, et al., 2019) the observations were made concerning the whole groups, not individuals. While the SCM can be applied for all levels (personal, group, national) that might be one the reasons for results observed as clothes serve as an indicator of status on an individual level while ethnicity is a group feature.

All ratings gravitated towards more neutral values, now all of the groups, though statistically different, are situated approximatelyat the middle point of both dimensions. That might be due to a sample collected and possibility will be explored further in the limitations part of this paper. This might also be partially explained by the fact that participants were not compensated for participation in any way and in the last decade it had became a common knowledge that people who participate in psychological research freely (without any compensation) are oftentimes more tolerant and less judgmental than general public. Having said that, it is important to mention that I received several email and a number of comments from people complaining that a person should not and cannot be evaluated based on their looks and that “all people are good” so some answers might have been compromised by social desirability bias leading participants to answer more positively.

The blend of ambivalent and unambivalent stereotype content goes somewhat logically with the fact that countries that are more unequal in income distribution tend to have more ambivalent stereotype content and vice versa (Durante & Fiske, 2017; Fiske et al., 2016; Fiske, 2017, 2018) and as of 2016 according to Organisation for EconomicCo-operation and Development Russia had an adequate level of income inequality of 0.33 (OECD, 2020). While most of these studies find correlational links, an experiment showed that only competence dimension is affected by the inequality - amid greater inequality high status targets are perceived as more competent and low status targets - as less competent then in relatively equal conditions (Connor et al., 2020)

Also worth noticing is the fact that, in the end, there were more questionnaires filled for Russian target(Russian - 208 > Uzbek - 183 > Chinese - 168 > Chechen- 159). The questionnaire was programmed in such a way that it assigns each participant one of 8 conditions based on participants sequential number which should result in an even number of participants in each of 8 groups. The discrepancy we see means that people began filling out the questionnaire and then left the survey, and much more people left the page when presented with a picture of Chechen (Chinese, Uzbek) target.

Conclusion