Vote estimate for PSOE and PP, obtained through ASEP’s model based on the answers of a national sample of 1,200 respondents representative of the Spanish population 18 years and over, demonstrates that there is an inverse and statistically significant relationship between the difference (in percent points) in vote estimate for these two political parties and estimated abstention derived from the same model, that is, that when estimated abstention is high the difference in vote estimate for PSOE and PP is small, and when estimated abstention is low the difference in vote estimate is larger. In fact, Pearson’s correlation coefficient between these two variables is r = .51 (significant at the .01 level). The time series since the last election in March 2004 shows that the difference between vote estimate for PSOE and vote estimate for PP was larger during the months immediately after that election (it reached 11 percent points in May 2004, when PSOE formed its government), something quite natural because of the effect of results in Rs answers, which tend to overestimate vote intentions for the winning party and to underestimate vote intentions towards the losing party. Thus, between March 2004 and May 2005 the difference was always favourable to PSOE, almost always above 5 percent points (the real difference in the 2004 elections being 3.8 percent points over 100 electors), but since June 2005 the difference has never been higher that 5 percent points. As a matter of fact, during the last sixteen months the difference has been larger than the real difference in the last elections only on four occasions, and even in two surveys (December 2005 and February 2006) the difference was favourable to PP, two points in the last of the two mentioned dates. In summary, it can be said that since June 2005 the difference in vote estimate for the PSOE and the PP has almost always been smaller than that really obtained in the 2004 elections, and the maximum fluctuation has varied from 4 points favourable to PSOE to 2 points favourable to PP, but in the majority of the other months the difference has varied between two and three points favourable to PSOE. It must be underlined that there is no observable trend maintained through time, but fluctuations around a difference of 4 points (the same, more or less, than that obtained in the 2004 elections), and with a estimated abstention of 23.2% (when that really observed in those elections was 22.8%). Taking into account, besides, that sampling error is 2.5, it may be said that vote estimate during the last year and a half suggests a tie situation between the two largest political parties, and in any case it shows a difference which is very similar to that which was obtained in the last 2004 elections, so that variations are mainly due to abstention estimate, in the sense that a larger abstention favours PP while a larger participation favours in general PSOE.

to enlarge image, click over it