Issue link: https://maltatoday.uberflip.com/i/1521974
3 maltatoday | THURSDAY • 6 JUNE 2024 SURVEY PN slips by one point points, support for third par- ties peaks at 9.7% and absten- tion peaks at 36%. A breakdown by education al- so suggests a sharp divide, with Labour leading by 16.8 points among the secondary educated and by a staggering 21.2 points among those with a primary level of education. However, Labour's lead drops to 11.2 points among those with a post-secondary level of education while the PN is lead- ing by 12.9 points among the tertiary educated. Support for third parties is also highest among the ter- tiary educated (10.3%) but decreases to 5.2% among the post-secondary educated and to just 3% among the second- ary educated. Third parties are practically inexistant among voters with a primary level of education. Labour dominates the south by wide margins For the first time in the cam- paign, the survey shows the PN enjoying a small 2.3-point lead in Gozo where Labour enjoyed a 5.9-point lead in the last sur- vey. The PN is also leading by 2.6 points in the Northern region which includes localities like Mellieha, St Paul's Bay and Mosta and by a more substan- tial 9.6 points in the North Harbour region which includes Sliema, Birkirkara and Msida. But while the PN has made small inroads in northern and central Malta, the PL is lead- ing by a very wide margin of 27 points in the South East, which includes localities like Marsas- kala and Zejtun, and a stagger- ing 33.1 points in the South Harbour, which includes the PL's Cottonera strongholds. But in the Western region the two main parties are practical- ly neck to neck as they were in previous surveys. Support for third parties also peaks at 10.6% in this region which includes Attard and Zebbug. Methodology The survey was carried out between 22 May 2024 and 4 June 2024 for which 1,007 peo- ple opted to complete the sur- vey. Stratified random sam- pling based on region, gender and age was used. A fraction of those who opted to complete the survey chose not to answer some of the questions for which they are treated as missing val- ues. Missing values analysis was then carried to determine the type of technique to replace the values. A combination of logistic and linear regression with predictive mean matching where applicable was used to replace certain missing values completing and enlarging the sample set from which the fi- nal results were extracted. The margin of error for this result 3.68% for a confidence interval of 95%. Voting intentions by education Voting intentions by region Voting intentions by gender Voting intentions by monthly income The survey suggests that increased political polarisation has not resulted in any overall increase in the turnout which has remained stable at 69.9%