Let Your Voice be HEARD.
Election Polling Methodology:
Pre-election polls are conducted to collect relevant and accurate opinions shared among voters. However, actual voters are an unknown population, which only exists on (or, with absentees, shortly before) Election Day. Pre-election polls make their best estimate of this population.
IAM polled the voting age population of the USVI because the registered voter population is similar to the demographics of the adult population of the USVI. Likeliness to vote and expected voter turnout are then determined by survey response.
As of June 6, 2022, the Election System of the Virgin Islands' voter data shows that the number of registered democrat voters is 34,444. To achieve the widely established confidence level of 95% with a 5% margin of error, an accurate sample size for the democrat voters in the USVI is 380 participants. With a democrat participants turnout of 752 (as of June 5, 2022), our margin of error for this Primary Election poll data is 3.53%. For the island of St. Croix, our margin of error is 5.07%, with a sample size of 365 participants. For the island of St. Thomas, our margin of error is 5.24%, with a sample size of 343 participants.
For the island of St. John, we included a population proportion of 3% to reflect the latest Democratic primary demographics from the USVI Elections office with respect to the number of registered voters of the other islands. St. John's sample threshold was calculated at 44, and the respondents received was also 44. We did not adjust the proportions for St. Croix, and St. Thomas since their proportions (48% and 49%) are similar to the default value (50%) for considering sample sizes. Therefore the MOE is +-5% for St. John.
We conducted (non-overlapping) online and phone surveys to achieve the optimum mix of online and traditional sampling methods for the USVI's target population of democrat voters. This approach mitigates any over or under-representation of any demographic within the survey sample.
Based on the statistical insignificance in the difference between weighted and non-weighted data, the target demographic of democrat voters fit within a negligible degree of error from the demographics of the respondents to the poll. Therefore, weighting the Primary Election results would not have resulted in any significant difference compared to the collected data.