It’s a bit exaggerated due to other factors (see below). It would be great to see a controlled study where some randomized participants were encouraged to switch to cycling to see those results.

“Due to the limitations of the SLS, we were unable to adjust for some potential confounders, such as income or body mass index at baseline, we were not able to account for time-varying confounding and excluded individuals with missing covariate data. Additionally, active commuting may be associated with other forms of physical activity that contribute to total physical activity levels.39 However, we were unable to adjust for other forms of physical activity, another limitation of the dataset used. This may have led to an overestimation of the effect estimates, particularly for cyclist commuters who have been shown to have higher levels of overall physical activity than other commuters.7 Nonetheless, previous research has demonstrated that beneficial associations between active travel and health remained after adjustment for other physical activity.3” https://bmjpublichealth.bmj.com/content/2/1/e001295#%3A~%3Atext=Due+to+the+limitations%2Cother+physical+activity.3

I dug into the survey and it appears they intentionally skewed the results by the word choice and order of the questions.

For example: “5. Do you [favor or oppose] the Biden administration approving taxpayer-funded weapons and other military support to Israel, even if the U.S. government has no control over whether the Israeli military uses those weapons on innocent civilians in Gaza, or are you undecided?”

The wording in that question will predispose a respondent to view the Biden administration negatively, even if that wasn’t the main reason that a respondent chose not to vote.

Worse, they asked that question before they asked about sitting out the election, predisposing the respondent to view that topic as more significant in their decision to sit out the election.

If the purpose of a poll was to figure out the reason for not voting, this is not a scientific way to do it. It predictably would skew results as it appears to have done.

It’s doubtable this poll has uncovered something that countless other post-election polls have missed.

It’s more likely that the pollster is biased. This is the first sentence of About section of the pollster’s website: “Palestinians, like all people, are entitled to live in freedom and with their human rights respected. Yet, for too long, the US government has funded and enabled Israel’s denial of Palestinian human rights.”