We use cookies in order to improve the quality and usability of the HSE website. More information about the use of cookies is available here, and the regulations on processing personal data can be found here. By continuing to use the site, you hereby confirm that you have been informed of the use of cookies by the HSE website and agree with our rules for processing personal data. You may disable cookies in your browser settings.

  • A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Svetlana Zhuchkova won Student Research Paper Competition

On February 12, the deputy head of the RSG Svetlana Zhuchkova and the head of RSG Alexey Rotmistrov were invited at the meeting of winners of Student Research Paper Competition (SRPC) with the rector of HSE Yaroslav Kuzminov. After the discussion between Mr. Kuzminov and students, an award ceremony was held. Svetlana was among the winners. Her paper was written under the guidance of Alexey Rotmistrov.


Svetlana took the 3rd place in the nomination “The best Student Research Paper in sociology for bachelors”. Her paper describes principles of handling missing values in CHAID. The content of the article directly correlates with the theme of the RSG’s studies, because it provides the reference to one of the two primary methods of the search of interactions between categorical variables - decision trees. The paper considers the unique feature of the CHAID - the opportunity to include missing values in a model without imputing. Although in literature this feature is defined only as an advantage, there is still no evidence of the correctness of including missing values into the analysis. Despite this, tree models with missing values can be found in empirical studies regularly.

The purpose of the paper was to find out if the CHAID defines missings to nodes correctly and what are the consequences of including missing values into the model. Using a series of statistic experiments, Svetlana discovered that the method identifies missings correctly. However, in the most cases including missing values is followed by changes in the tree structure. Therefore there is a risk of getting substantially incorrect conclusions.