About RockNet
RockNet Home PageRockNet Site Map
Rockville City Elections 1999
Election Analysis

Alternatively, you may also view this document by selecting one of the following files.
      el99anal.doc (Microsoft Word 97 Document, 153KB)
      el99anal.pdf (Adobe Portable Document Format, 150KB)
The vote tallies and calculations used in this report are available in the following file.
      el99anal.xls (Microsoft Excel 97 Spreadsheet, 52KB)
Note: All vote tallies are from official election results.

Roald Schrack

November 22, 1999


The following table shows the number of ballots cast and registration in the last several elections.

Year Ballots Registration B/R Previous Voters New Registration
1991 5451 20,648 .26 7916 1741
1993 1449 23,077 .06 9889 4409
1995 4323 22,787 .19 8133 1317
1997 5211 22,660 .22 6968 3299
1999 4690 22,676 .21 6898 1451

Although there are only small differences in the voting between 1997 and 1999 a few of things were different in the two elections. In 1997 the incumbent Mayor and all 4 incumbent Council members ran and there was one Council challenger while in 1999 there was an open seat on the council with 6 council candidates, three of them incumbents. Election day in 1997 was mild and clear, while heavy rain persisted from noon until after 4:00 p.m. on election day 1999. These two differences had opposite effects on the turnout.

The effect of the rain can be estimated from the relative number of absentee ballots and polling place ballots. The following table shows absentee ballots and their ratio to the polling place ballots for the last several elections.

Year Absentee Absentee/Ballots Polling Place
1989 175 .028
1991 168 .032
1993 48 .033
1995 180 .042
1997 156 .031
1999 171 .038

The average of Absentee Ballots/Polling Place Ballots for the last four non-rainy years is .031, the ratio for 1995 is relatively high because the weather was rainy for most of the day. If the same fraction of the voters voted by absentee ballot then the implication is that the rain decreased the number of polling place voters by .31/.38= .82. If it had been a clear day, 4519(polling place ballots)/.82= 5512 would have made it to the polls making the voter participation 0.25 instead of 0.21. Figure 1 shows how weather and number of council candidates determine the voter participation.

fig1.gif (10386 bytes)

A record was kept of the number of ballots cast by 9:00am and by 4:00pm. Figure 2 shows the fraction of the total vote at a district polling place for the two times. Note that the variation is not very great at 9:00 am but is much greater at 4:00 p.m. Using this data it was possible to predict the final total vote with 2% accuracy at 4:00 p.m. based on similar records taken in 1997.

fig2.gif (14241 bytes)

There are 12 voting districts in Rockville placed so that no voter has to travel far to his/her polling place. The Districts have very different numbers of registered voters and thus they have a different impact on the election. Figure 3 shows the total number of registered voters for each district for the 1999 election. The way that the districts have been chosen has led to an interesting side effect. Since the districts also reflect varying demographic characteristics and economic factors, the voters make different political decisions. In addition, changes in registration and voting reflect changes in the characteristics of a neighborhood.

fig3.gif (24555 bytes)

Figure 4. Shows the change in the number of registered voters from 1997 to 1999. The 6% increase in registration in District 2 is too large to be a random statistical fluctuation.

The growth in registration in district 2 is due in large part to the construction of new homes near Dogwood Park and the reconstruction of the Blandford Street apartments. In other districts the addition of new residents was offset by the departure of the previous residents of the existing housing.

fig4.gif (12827 bytes)

District Comparisons

Figure 5 shows the total vote in each district in the 1999 election. Note that the vote varies by almost a factor of four. Comparison to Figure 3 will show that districts with low total vote may have higher than average participation (ballots/ registered voters). This comparison is shown explicitly in Figure 6.

fig5.gif (27287 bytes)

In Figure 6, the relative voter participation is shown for 1997 and 1999 elections. Note that the participation in districts 5, 8, 9, and 12 dropped. This is probably related to the support in those districts for Mr. Haughey in the 1997 election. Note that District 11 has a consistently high participation. District 11 has the New Mark Commons subdivision in it.

fig6.gif (14122 bytes)

If New Mark Commons is removed from the district the participation would be about 0.17, similar to districts 7, 8, 9, or 10. Participation by New Mark Commons residents is a remarkable 0.72. This no doubt related to the fact that the Mayor is also the Administrator of the New Mark Commons Homeowner's Association. It will be interesting to see how participation is affected when the Mayor is no longer a candidate for elected office in Rockville. It is, of course, a great tribute to the Mayor that those who know best of her administrative ability are the most enthusiastic about her candidacy.


The coverage of the race by the Rockville Gazette was quite good with separate articles on the different candidates and a voters guide type insert the week before the election. The city cable channel gave time to all candidates for a short statement and the city newsletter had a n election insert that reproduced the candidate statements. The Alliance of Rockville Citizens issued a newsletter which gave uniform coverage to all candidates. The newsletter was reproduced on Rocknet, the community internet page. In addition there were four community forums where the candidates could be questioned directly by the public. The electorate was well informed about the candidates from a multitude of sources.

Figure 7 shows the fraction of the total votes in a district received by each candidate. Rose Krasnow, the candidate for mayor, clearly led the race, receiving more votes than any council candidate in every district. The support for her opposition, Lih Young dropped from 619 votes in 1997 to 387 in 1999. The candidates that ran with Krasnow on the Committee for Rockville's Future slate ( Bob Dorsey, Glennon Harrison, Anne Robbins, and Bob Wright) received from 69% to 74% of the vote as compared to Krasnow's 87% of the vote. Charlie Adams and Eric Wang, newcomers to Rockville politics , trailed with 48% of the voters giving one of their four council ballots to Adams and 19% to Wang. As might be expected , there was a strong anti-correlation to the vote for Robbins and the vote for Adams. Where the graph is hard to read please refer to the tabulated results.

fig7.gif (18448 bytes)

Not shown in Figure 7 is the number of blank ballots. In the race for Mayor there were only 206 blank ballots out of a total of 4690 or 4%. In the Council race each voter has the right to make four selections. In the 1997 election there was a great deal of "bullet balloting" where a voter chose to use only one of his council votes in an attempt to boost the chance of election for a single council candidate. In that race the missing ballot fraction was 1.3 in District 12. This result could be achieved if about half the voters in District 12 voted for only one council candidate. Note that again in

fig8.gif (13668 bytes)

the 1999 race that district 12 has the largest missing ballot fraction. There does not seem to be a strong correlation of the missing ballot fraction with any particular candidate in the rest of the districts and as far as is known, none of the candidates asked their supporters to "bullet ballot" for them.

Because of the rain fewer votes were cast in 1999 than in the 1997 election. The incumbents obtained a greater fraction of voter support than they did in 1997. Figure 9 shows the gains made by the incumbents over their support fraction in the 1997 election.

Averaged over all districts the relative gains were:

Krasnow 1.05
Dorsey 1.01
Harrison .98
Wright 1.07

fig9.gif (16019 bytes)


The district correlation analysis shows the degree to which the pattern of votes for the candidates in the 12 districts is similar. Some candidates have particularly strong support in a particular district that is not shared by the other candidates, this reduces their correlation coefficient. The candidates whose vote patterns are most similar are Krasnow and Wright. The correlation values for Young and Wang are not included because they were they were so low. Only the upper half of the table is shown, the lower half is identical.

  Krasnow Adams Dorsey Harrison Robbins Wright
Krasnow 1 0 .42 .54 .51 .77
Adams   1 0 0 0 0
Dorsey     1 .71 .20 .71
Harrison       1 .25 .62
Robbins         1 .48
Wright           1

Charlie Adams vote pattern was anti-correlated with all other candidates. His supporters , in general, were distributed in a different pattern in the city than those he ran against. The following table shows the anti-correlation values for Adams in relation to the other candidates.

Krasnow Dorsey Harrison Robbins Wright
.58 .60 .54 .71 .65

The slightly higher anti-correlation for Robbins indicates that Adams supporters were less likely to give one of their 3 remaining votes to Robbins than one of the incumbents.