Date: Tue, 18 Oct 1994 15:48:49 -0400 From: BRIAN VARGUS To: Multiple recipients of list Subject: Re: Undecideds in congressional election polls ----------------------------Original message---------------------------- To Whom It May Concern: We have just completed a statewide survey for the Indianapolis Star and WISH-TV of 1061 registered voters in Indiana. We have high undecideds in all races be cause the question we ask is "open-ended," i.e., "Who do you intend to vote for in the race for Congress in your district?" If they say "don't know" or "unde cided" we then read them the names of each candidate. The lowest DK is 35% in the 8th Congressional District. Statewide th4e DK and NA total 38.8%. Just for a taste of the "spiral of silence" idea, I crosstabluated the Congress ional vote choice by party by announced approval of the President. The followi ng is what we found. APPROVAL RATING OF CLINTON Approve Disapprove DK NA CHOICE FOR CONGRESS Democrat 207 112 20 4 Republican 58 220 11 4 Other 6 4 3 0 Don't Know 140 173 30 5 NA/Refuse 25 27 8 4 Again, this is a statewide sample of registered voters. Note that while 40% of those that approve of the President are in the DK on candidate choice, almost 50% of those that disapprove are DK on candidate choice. Is this some e vidence of a small spiral of silence effect? At this point we are running the same table within party ID. A secondary note on races: Senate (1061 registered voters) Lugar (R) 49.5% Jontz (D) 15.9 DK/NA 34.6 2nd Congressional Dist (open) (538 Registered Voters) Hogsett (D) 30.3% Macintosh (R) 26.2 DK/NA 43.6 7th Congressional Dist (509 Registered Voters) Myers (R)(inc) 39.9% Harmless (D) 19.8 DK/NA 40.3 All these are panel studies so we have the same people at least twice. The DK/ NA categories have stayed at low 40's since mid-September. If you use this dat a in any public way please attribute it to the Indianapolis Star-WISH-TV. We have used the same question format in election panels for 12 years and the DK/NA tend not to move much even when we go back the weekend before the election. All interpretation welcome! Brian Vargus, IUPUI Public Opinion Lab.