Graphic Analysis

American Soviet Russian "Our"/Sov/Rus French Italian European Indian German Foreign Other 20% 11% 9% 1,5% 12% 7% 6% 3% 3% 10% 17,5% References to National Cinemas in Interviews: Percent of Total References

As the graph shows, references to American cinema occur most frequently in the interviews, as often as references to Russian and Soviet cinema combined (“our Cinema” refers to either Russian or Soviet cinema). See the detailed analysis under Qualitative Analysis.

100% art adult action comedy wonder tale war film romcom horror films for children silent films References to film genres (% by gender) red = female viewers, blue = male viewers 100% = all interviews by 'f' or 'm' 67% 67% 67% 67% 83% 83% 50% 33% 33% 33% 33%

Genre references by gender of interviewees. We divided interviews into two groups (female and male interviewees) and analyzed the frequency of mentioning of specific genres by female and male interviewees.

For our analysis we used collection of interviews on Obdurodon server and xquery FLOWR expression

  1. xquery version "3.0";
  2. let $index := doc('/db/course/rmtp/index.xml')
  3. let $interviews := collection('/db/course/rmtp/data')/interview[.//interviewee/@ref = $index//person[.//gender/text()='f']/@xml:id]
  4. let $distinctGenres := distinct-values($interviews//genre/@ref)
  5. for $distinctGenre in $distinctGenres let $countInterviews := count($interviews//body[.//genre/@ref = $distinctGenre])
  6. where $countInterviews gt 1
  7. return concat($distinctGenre, " occurs in ", $countInterviews, "interview(s).")

We used the data to build bar graphs representing how many interviews by female and male interviewees mention genres more than once in the interviews. There are two ways to read these bar graphs. First, we can look at the points of differences and convergence between genre preferences. Both male and female interviewees favor comedy as their favorite genre. The rest of the favorite genres differ. Second we can examine the range of genres that male and female interviewees favor. The range of genres discussed by female interviewees is almost twice broader.

100% Use of Pronouns in Interviews by Gender (%) Red represents female interviewees and blue represents male interviewees. Interviews are weighted. Total F:62%/M:38% Plural F:73%/M:27% Plur, Poss F:71%/M:29% Plur, Pers F:73%/M:27% Sing, Poss F:52%/M:48% Sing, Pers F:58%/M:42%

The graph visualizes the use of pronouns in the interviews. In particular, we focused on the use of first person pronouns (“I” and “we”) and possessive pronouns (“my” and “our”). While female and male interviewees used roughly the same number of “I” and “my” (52/48 and 58/42 respectively), female interviewees used significantly more plural pronouns, both personal and possessive, than male interviewees. Women used 71% of all “our” pronouns and 73% of all “we” pronouns. For a detailed analysis see Qualitative Analysis.

100% Use of References to Community as a National Community in Interviews by Gender (%) Red represents female interviewees and blue represents male interviewees. Interviews are weighted. Total F:76%/M:24%
100% Use of Hesitation in Interviews by Gender (%) Red represents female interviewees and blue represents male interviewees. Interviews are weighted. Total F:59%/M:41% Lexical hesitation F:52%/M:48% Pauses F:66%/M:34%

Female and male interviewees used a roughly equal average number of lexical means of hesitation: ну, наверно and может быть (well, perhaps, may be) . The difference in the use of non-lexical hesitation (pauses) between genders is much more pronounced: women paused in their answers almost twice as often as men. See further discussion of the graph under Qualitative Analysis.