The results of our August Poll is in and well, there was quite some news. The national poll was conducted using a combination of phone calls (live caller) and online surveys from August 22nd to September 5. A total of 516 responses were collected across all 16 regions and demographic groups.
- The poll used a combination of live phone calls and internet surveys. This was to ensure the sample was as representative of the voting populace as possible.
- A total of 516 respondents were recorded, with a margin of error of +/- 4.
- Statistical adjustments were made to account for variations in the following key demographic vectors: age, gender, education level and region of residence.
- While 516 respondents may seem small in the face of 16 Million plus registered voters, the sample size is still statistically valid within the margin of error.
- By comparison, most polls in the US register about 1,000 respondents only despite the fact that the voting age population in the US is upwards of 250 Million.
- Yes, sample size does matter but only up to a point. Once you are past the point of statistical significance, you start to see the same movement in the data with each new response that comes in. The most important thing then is to collect enough respondents to validate statistical significance and to ensure the sample itself is as representative of the voting public as is possible. Our August poll met both criteria.
- As to the results of the poll, there were some notable surprises. Nana Addo won a plurality of support but not a majority while a fairly large share of voters remain undecided. Get all the details here.