ITCHS Home        HQN Contents         Permissions



The Humor Quotient Newsletter

Vol. 8, No. 2, March 2002, Winona, MN


Mental Humor and Preference among American Patriotic Songs II


[The last issue of HQN was devoted to correlations between the four mental humor types tested in HQT and relative preference for four patriotic songs in the wake of the terrorist attacks of September 11, 2001.  This issue continues to investigate that database—Ed.]


Data from the Humor Quotient Test is typically considered at two levels: an analytic level that considers each of four sub-types of humor of the mind against other variables and a synthetic level that considers groups of two of the four sub-types against other variables.  After the September 11 attacks on the United States, students in three sections at Winona State University provided participants for a study of humor preferences compared to preferences for one of four patriotic songs: “The Star-Spangled Banner,” “America the Beautiful,” “God Bless America,”  and  “Amazing  Grace.”  The Patriotic Song Side-Test came in two parts.  The first was a reprinting of the patriotically-usual stanzas of the four hymns accompanied by an instruction to reread the lyrics meditatively.  The second page then asked participants to divide 10 points among the four songs for each of five questions about appropriate venues for patriotic song performance.  Thus, we could expect that participants had been very recently reminded of the exact wording of the most patriotic stanzas of all four songs before deciding where which songs would be most appropriately performed.





The impetus for this research was journalistic controversy about a proper national anthem, perhaps most strongly argued by Ellen Goodman, who proposed replacing a bellicose “Star-Spangled Banner” with a more neutral and appropriate “America the Beautiful.”


As we noted in the last issue, from an analytic standpoint, the experiment produced high-confidence results that the relative appreciation of the various songs was related to humor preferences. Notably, “The Star-Spangled Banner” was strongly correlated to Incongruity and Word Play while “God Bless America” was negatively correlated to these.  “Amazing Grace” was strongly correlated to Gotcha and Incongruity humor while “America the Beautiful” was not strongly correlated to any of the four analytic forms of mental humor.



Now we can take such analytic results and assume that they stand proxy for synthetic responses, that is, for example, that the strong correlation of “Star-Spangled Banner” with Incongruity and Word play is also a strong correlation between “Star-Spangled Banner” and the synthetic construct of Intellectual (=Incongruity + Word Play subscores).  But it is possible that the synthetic math would not work out as expected, especially if, for example, some psychological law biased people to like Gotcha if they first liked Incongruity.


We thus continue our investigation with slope coefficients of three humor personality types against the four patriotic songs.  (See Table 1)  There are six synthetic humor types, but the six consist of three pairs of mathematical opposites.  Thus to be anti-Crusader is to be pro-Consoler, to be anti Advocate is to be pro-Reconciler, to be anti-Bridgebuilder is to be pro-Intellectual.  So negative slope coefficients can be read as positive correlations with the opposite pole of the pair.



Table 1: Synthetic Humor Personality/Patriotic Song Preference Slope Coefficients


                                                CRU                      ADV              BRI


“Star-Spangled Banner”         -.0294                   -.0055            -.1551


“America the Beautiful”          .0836                   -.0594             .0037


“God Bless America”            -.1427                   .0223             .2436


“Amazing Grace”                    .1432                    .0549            -.0424






Four of the slope coefficients from Table 1 stand out as greater than .14 in absolute magnitude, averaging .1712.  The other eight average .0308, with two scores under .0100, in other words, virtually random.


None of the correlations with Advocate measure greater than .0594, whereas the correlation between Bridgebuilder and “God Bless America” is the highest of the set at .2436.


“America the Beautiful” has no humor correlation above .0836, while “God Bless America” is highly correlated with Bridgebuilder (.2436) and highly correlated with the opposite pole of Crusader, that is with Consoler (-.1427).


It should be noted that none of these slope correlations speak to the relative popularity of the four songs or to Goodman’s proposal to replace “The Star-Spangled Banner” with “America the Beautiful.”  The relative popularity of the four patriotic songs among participant responses was:


Table 2: Relative Popularity of Four Patriotic Songs By Gender


                                                                     Total              Female             Male


“The Star-Spangled Banner”                       1287                767                  450


“America the Beautiful”                                777                505                  202


“God Bless America”                                    819                540                  279


“Amazing Grace”                                          641                442                  199


Probably not surprisingly, then, “The Star-Spangled Banner” was in no danger among our respondents from Goodman’s suggestion of replacement by “America the Beautiful,” and in fact, “America the Beautiful” came in third behind “God Bless America.”  Probably also not surprisingly, “Amazing Grace” trailed the pack, but it is perhaps surprising that it did not trail that much behind second and third place.


Thus, for all respondents strong correlations existed between “God Bless America” and both Bridgebuilder and Consoler; between “Star-Spangled Banner” and Intellectual; and between “Amazing Grace” and Crusader.  In this synthetic analysis, the only result not particularly indicated by an analytic analysis was the correlation of “God Bless America” with Consoler.

HQN Editor:  Paul Grawe, Institute for Travesty, Comedy, and Humor Studies, 678 Sioux Street, Winona, MN 55987. Tel: 507-454-4141. email:





In the first part of this report, it was argued that the ubiquitous God Bless America signs after 9-11 thus probably showed Americans bonding together, bridging all internal divisions, as a first step in the war against terrorism.  This result of course closely parallels general polling indications of support for the war and for the President.  Both results show absolutely extraordinary unanimity across all social classes in the United States.


But the synthetic analysis also shows that God Bless America signs may have had a consoling quality as well, Americans consoling Americans on the largest single-day casualties of American history and moreover casualties entirely on American soil.


The HQT testing program at WSU regularly analyses gender differences because such differences are so often significant.  In an age of women’s studies, it seems highly inappropriate to report general findings without determining whether these findings are male-skewed.  At the same time, because WSU is associated with Mayo Clinic in its nursing program and in other ways draws significantly more female than male students, WSU databases typically are skewed in favor of female respondents.  Correlations between humor type and patriotic song preference within gender provide some additional insights.  Table 3 provides Female and Male slope coefficients, and these tend to suggest that men’s humor is more related to patriotic songs than is women’s.  Also, 4 of 6 “America the Beautiful” results are suddenly higher than .14, whereas no such combined-gender high-slope results were found.


Table 3: Humor Type/Patriotic Preference Slope Correlations, By Gender


                                                           Female                                   Male


                                              CRU      ADV      BRI            CRU      ADV      BRI


“Star-Spangled Banner”      -.0203      .0000    -.0964         -.0588   -.0382    -.2997


“America the Beautiful”       .1628     -.0317   -.0336          -.1855   -.1711     .1488


“God Bless America”          -.1540       .0068    .1950          -.0765    .1234     .4298


“Amazing Grace”                  .0833      .0249    -.0575          .4362     .2419    -.0504


N.B. that “America the Beautiful” is strongly correlated to Crusader humor for women and strongly correlated against Crusader and thus toward Consoler humor for men.


Paul Grawe

Winona State University




ITCHS Home        HQN Contents         Permissions