This is a ridiculous claim disproven daily by thousands of daycares practicing color-coded stable groups.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3513287/
It likely also depends on how prominent the color coding is. Just a marker would probably be too abstract for children to be factored in. But I'd question your claim if children are clad differently in an obvious fashion like different shirt colors f.x.
The study you posted does not describe children bullying or forming subgroups based on wearing a color shirt.