There is a place on Islington Street in Portsmouth New Hampshire where there is a popular tea and coffee shop, White Heron. At this particular location the Parking and Traffic Safety Committee is considering adding a crosswalk, which requires a bump-out from the curb to make it clear there is only 1 lane there (there is room for 2 cars and cars often make it two lanes with the left going straight and the right turning at the next intersection). This causes traffic to back up a little more during busy times as that "extra lane" is shortened and during busy times cars must wait until they get past the bump-out to begin the 2nd lane. It also makes it so bikes need to merge into that lane to avoid the bump out. The planning committee has temporarily tried this 2 times before, the first time had many problems and after feedback, they changed a few things (these are not important for this post) and tried again. The 2nd time was more successful, but still saw complaints. Both times there was an online comment period for people to provide feedback. A few days ago, this temporary crosswalk and bump-out returned for another 1-month trial, this time during the busiest traffic time of the year, our touristy summers. It is also my understanding that to have the crosswalk, the bump-out, or something similar, is required as a crosswalk can only cross 3+ lanes of traffic if there is a stop sign or traffic light.
![]() |
| A photo of the first temporary crosswalk. The current set up looks quite different. Photo Credit: Wired Article |
In a large local Facebook group, a member started a thread complaining that it had returned and many group members commented. There were many comments claiming that nobody wants the cross walk and why did they provide feedback if it was just going to be ignored. It seemed clear that that most passionate of comments were of the thought that this is crosswalk is a bad idea, it is obvious it is a bad idea, and the city is ignoring the people and catering to some business. There were statements that only those who frequent the White Heron are in favor. There were many other things said and some really good points on both sides, but that is not what this post is about, this post is about answering a few simple questions:
1) Are members of the group generally for or against the crosswalk?
2) What is the difference in crosswalk opinions for those who live near the crosswalk vs those who do not vs those who only work in the area (and live in a different city)?
3) What is the difference in crosswalk opinions for those who frequent White Heron vs those do not?
4) What is the difference in crosswalk opinions for those who drive through the proposed crosswalk daily vs those who do not.
To answer these questions I posted a survey to the group and promoted it. I received 104 responses. I asked if they were strongly against, somewhat against, strongly for, somewhat for, or neither for or against the crosswalk. I asked where they lived, if how often the patronize White Heron, how often they cross at the spot anyway (I used the word illegally and one survey taker commented that it is not actually illegal to cross there as it is an intersection and pedestrians have the right of way at intersections), how often they drive by that spot and if they patronize other businesses near White Heron. Here are the summarized results for each question and the raw data: Results and Raw Data.
This is what I have found.
Here are the results for the general question of people being against or in favor of the crosswalk. I combined strongly and somewhat for against and for into a single category. In this survey, people were in favor at a rate nearly 2 to 1 versus those against as 55% were in favor at 30% were against.
Perhaps it matters where people live. Many commented that only those who live near there would be in favor, so do we see a difference when we condition on location? There appears to be no difference as these distributions are similar to the overall:
Next, we can look at how patronizing White Heron affects opinions. For this, I combined those who are frequent (once or twice a week) and those who are very frequent (a very regular customer) into one group called frequent, those who have never been there or only been there once or twice ever into a group called rarely/never and the rest into infrequent customers. One would expect White Heron customers to welcome this (in fact they signed a petition to make the cross walk permanent), so that is not what we are looking at. We are most interested in those who do not patronize White Heron and to a lesser degree those who do so infrequently. Here is where those against start to group:
Those who do not patronize White Heron are against the crosswalk 2 to 1, while those who go frequently are in favor 3 to 1. Interestingly, those who go to White Heron infrequently (a couple times a month) are also in favor by almost 3 to 1 -- these are also our largest group.
If it makes sense that those who are impacted positively when access to White Heron is improved overwhelmingly support, one might also hypothesize that those most inconvenienced, those who drive by often would be overwhelmingly against. However, it is not that simple as there is some bias in this data as those who frequent White Heron are also likely to drive by often. Regardless, it is interesting to look at:
For this, I combined those who drive by daily with those who drive by more than daily (daily and often multiple times in a day) as a yes. More are in favor than not, but it is not close to the 2 to 1 margin we have seen early. Unsurprisingly, those who do not drive by daily like the crosswalk at a 3 to 1 clip. What if we break down those who drive by daily by the frequent vs the very frequent:
We can see that those who are very frequent make up the majority of those against. Again, do not be fooled by how many are for, it is important that it is nearly the same amount, but there is some bias here as there are members of this sample who are also in the frequent White Heron customer sample.
Finally, there are those who are already crossing at that spot -- which according to one survey taker is legal and not jaywalking because it is an intersection (Islington and Albany). I am not sure that is true, but it would not surprise me, New Hampshire has many laws like this. For the record, I did phrase it as "illegally cross" when I asked. Once again, not a surprise, but interesting to see:
Conclusion:
I am not writing this to help support my opinion, rather to lay out some data so that I can generally be more informed. I think it is clear that many against commenters made assumptions that everyone shared their opinion and I think the data shows that is not true.
I am happy to answer a question you have, of if you want me to look for something else in the data (I could go way deeper, but this took me a couple hours to write, so I am only going as deep as this on my own). I will say that I wish I had asked if the participant biked in the area, that has been a big topic of discussion.
I hope you enjoyed.
-Stats Sam
Edit:
A comment pointed out that there are other establishments besides White Heron at the proposed location and asked me about it, here is how that looks:
I also was asked if there was a way to remove the frequent White Heron patrons from those who drive through the area often, so I did that as well:









Interesting and thoughtful analysis. Two observations I would add:
ReplyDeleteFirst, as i am sure you know, sample bias greatly favors those who frequent White Heron or wbo live nearby (me in both cases) as they are more interested in the outcome and likely to respond. In both groups, people that want a crosswalk may be more or less likely to respond to a survey. We dont know. So while the data shoe that living nearby does not correlate (assuming adequate sample size), frequenting white heron does. This suggests the numbers for or against are not representative of the general population unless you try to extrapolate to the results to the general population. But because we don't know whether people who respond to surveys differ from those who do not, we really can't reliably say much about public opinion at all based on this data. It is very interesting and thoughtful, but we have to be very careful about taking this type of data at face value.
My second point is traffic safety and street design are not popularity contests and should never be. Despite this, I suspect that the city's allocation of transportation funding is based more on addressing the concerns of the wealthy or developers who are more invested in local government than the general population may be. That bothers me but I don't have a good answer on what to do about it, other than communicate with local officials and hope they listen. Sometimes that feels even more dangerous or futile than crossing islington street at 5 pm in February when the sun angle is low.
Yes, non-response bias is always a consideration in volunteer surveys. It is almost expected. Recall that I am not trying to say this is what Portsmouth thinks, rather I the data is meant to speak towards what this online group thought. Perhaps I was not clear enough in my write up, but there is a clear correlation between those who drive by often (regardless of how often the patronize WH) and against as well. Further, my motivation was to inform myself on what the group actually thought, in fact when I first did it I expected to see many more against (as they were the most passionate and loudest in the discussion). I thought the against would be the most motivated to answer and so this data shows that either I was wrong in that assumption or there are few against that it appears through all the noise.
DeleteAs to your 2nd point, my survey was not about "hey lets see what people think and go with that" rather it was a "people are acting like everyone told the city that this was a bad idea and they are ignoring the people ... so lets see if that is true". In other words, I was just trying to open the eyes of people to how prevalent the other opinion was.
One observation: White Heron is not the only business in its building. There is also a Martial Arts studio in the building, which is frequented by many children. Some parents drive their children, but others (me included) often bike or walk our kids to the studio. When this crosswalk has appeared, I have been very, very grateful for it!
ReplyDeleteI have added this in at the end for you.
DeleteMy two cents: I drive by car at least 2 times a day, and frequent White Heron occasionally. It is a busy part of Islington, close to the light from either direction. People darting across the street is dangerous and a crosswalk would be helpful for that. Or they should be forced to cross at a slightly different spot, not in front of White Heron. The other problem is that White Heron at their peak has parking limits, thus adding to the congestion and potential pedestrian hazard. I find the crosswalk as a driver to be odd and dangerous, but I certainly can see the new one with the "markers." I'm concerned about people trying to cross there -- especially the kids, and there is no place to park on the White Heron side.
ReplyDeleteSomeone who admits to seeing both sides, a rarity.
Delete