### Table 2: Estimates of General Model with Mean Party Placements: Roughly speaking, ^ v 0,

"... In PAGE 9: ... Whether cij is in fact constant over i or not, replacing it with cj has important side e ects for estimation, a topic to which we now turn. Table2 repeats the regression in Table 1 with these mean party placements. The results for this table are qualitatively the same, except that estimates of c are now much less precisely estimated, more variable across issues, and overall less clearly distinguishable from zero.... In PAGE 9: ... Nevertheless, with this methodological adjustment, the results now provide somewhat greater support for the directional theory of voting | the restriction that c = v = 0 is more consistent with the data. [ Table2 about here.] MRL also worry that there might be omitted party-speci c attributes that a ect feeling ther- mometer scores, a standard statistical problem.... ..."

### Table 1: Speaking rates of broadcast news speech

2000

"... In PAGE 2: ... So, the experiments here roughly correspond to the partitioned evaluation (PE) with the baseline broadcast condition (F0) in the 1996 Hub-4 test [11]. Table1 shows the average speaking rates of the radio news from respective sources. It can be found that, on average, there are 5.... ..."

Cited by 12

### Table 1: Estimates of General Model: Roughly speaking, ^ v 0, ^ c gt; 0, and ^ 2 gt; 0. For this table, and all that follow, N = 8; 833.

"... In PAGE 8: ... Continuing the notation developed above, denoting the feeling thermometer scores of voter i evaluating candidate j as Tij and indexing each of the issues by k, we estimate the following equation by least squares: Tij = ? 6 X k=1 vkv2 ik ? 6 X k=1 ckc2 ijk + 6 X k=1 2k(2vikcijk) + ij (4) With six issues, we have six vs, cs and 2s to estimate | one for each corresponding v2 i , c2 ij, and (2vicij). The results, which appear in Table1 , indicate that v is usually fairly close to zero, but c and 2 are both large and positive.11 Thus, before correcting any methodological problems, the results do not provide unambiguous support for either the directional model (which requires v = c = 0 and 2 gt; 0) or the proximity model (which requires v = c = 2 gt; 0).... In PAGE 8: ...11 Thus, before correcting any methodological problems, the results do not provide unambiguous support for either the directional model (which requires v = c = 0 and 2 gt; 0) or the proximity model (which requires v = c = 2 gt; 0). [ Table1 about here.] If the results in table 1 are so mixed, how are Westholm and MRL able to draw such strong and opposing conclusions? The answer depends critically on their di ering methodological assumptions and resulting statistical corrections, a subject to which we now turn.... In PAGE 9: ... Whether cij is in fact constant over i or not, replacing it with cj has important side e ects for estimation, a topic to which we now turn. Table 2 repeats the regression in Table1 with these mean party placements. The results for this table are qualitatively the same, except that estimates of c are now much less precisely estimated, more variable across issues, and overall less clearly distinguishable from zero.... ..."

### Tables. INTRODUCTION

### Table of contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2001

Cited by 61

### Table 2. Speaking Rate Measures

1996

"... In PAGE 3: ...4. Prosody: Speaking Rate Descriptive statistics on speaking rate (words per second and syllables per second) are listed in Table2 . There were no significant effects of sex or dialect or their interaction on... ..."

Cited by 4