Studies on the opinions of cyclists about different street conditions.

Noland on Perceived Risk; Antonakos on Preferences; Sorton on Stress Levels (1st); Sorton on Stress Levels (2nd); Davis on Evaluation of Conditions; Epperson on Suitability; Landis on Hazard Estimation

home                 facilities

horizontal rule

1 Noland, Robert B.; Perceived Risk and Modal Choice: Risk Compensation in Transportation Systems; Accident Analysis and Prevention, Vol 27 No 4 pp 503-521, 1995

1.1 Description

The first part of this paper is a survey of various theories of risk compensation and analyses of them. These theories all consider that when an activity (highway use is the only activity that these studies consider) is made less dangerous its participants increase their intensity of participation: taking more risks, participating more frequently, etc. This is a controversial subject about which studies, conclusions, and opinions differ. Some bicycle advocates have taken these theories to heart, maintaining that when motoring is made safer for motorists, they drive more "dangerously" and the non-motorized suffer an increased accident rate. A comparable argument for cycling is that two-wheel brakes do not produce a lower accident rate than do coaster brakes; they merely allow the cyclist on a descent to get closer to the curve before braking. (I will not discuss the merits of this argument here.) One problem with these theories is that there may be a considerable difference between the actual change in "safety" and the perception of that change. The participants in the activity may either overestimate or underestimate the actual change in safety and thereby take inappropriate actions. Noland attempts to circumvent these problems by relying only on subjective evaluations of risk to predict what changes people might make in their choice of commuting mode if they changed their opinions about the relative risks of the various choices.

Noland made a mail survey of people who could commute to Philadelphia by transit, a survey that included a large number of cyclists. He asked each subject for his estimate of the probability of incurring an accident (in the next five years) and its probable severity if he commuted to work by each of the four modes: automobile, bicycle, transit, and walking. The answers had to be filtered to remove the unrealistic answers, such as the very high probability of having an accident while walking if the walking took 10 hours a day. Noland used a multinomial logit model, for which he had to devise and validate many coefficients. He also used varied factors to show that the comparative results were similar over a range of factors, thus suggesting that poor estimation of the factors is unlikely to have significantly affected the results. I cannot comment on the validity of these processes.

In general, most models rated the order of the risks of the modes (least to most) as transit, automobile, walking, and cycling. For a smaller number of models, the order was transit, walking, automobile, and bicycle. Noland then calculated the changes in modal proportions if the participant felt that there was a change in risk. Of most interest to us, he concluded that if people perceive that cycling has become 1% less risky, the increase in cycling will be somewhat more than 1%. This amount of change is far more than for any of the other modes. And he concluded that most of those who would change to cycling would come from those who had been walking. He writes that if we install many bike paths and thereby attract many people to cycling, but the urban bike paths turn out to be more dangerous than normal roadways, then we will have an increased cycling population operating at a higher accident rate. Noland also concludes that the largest effect of an increase in automobile safety would be a reduction of transit ridership.

1.2 Evaluation

Noland's data showing that most people think of cycling as the most dangerous mode agrees with other surveys. Whether cycling or walking is the more dangerous depends on whether one considers time or distance. The same has been said of the comparison of cycling and motoring. However, I disagree with Noland's conclusion that the greatest number of new cyclists would come from present walkers. That may be true in an area where there are many walkers, but not in the typical American city. I think that the mode most jeopardized by an increase in cycling is short-distance transit, for it is in that functional area that cycling shows the greatest competitive advantage once people realize that cycling is not as dangerous as they thought. However, I think that the greatest number of new cyclists would come from motorists, because there are an enormous number of trips made by car that could be done by bicycle. I think that Noland overestimates the effect of risk in modal choice for most modes and underestimates the effect of the other factors. The exception may be cycling, where, in my opinion, the public has a far greater overestimation of the dangers than it has for other modes.

2 Antonakos, Cathy L.; Environmental and Travel Preferences of Cyclists; TRB Transportation Research Record 1438

2.1 Description

Antonakos questioned cyclists participating in several tours in Michigan regarding their preferences in cycling conditions, their choice of mode for other trips, etc. Many of these cyclists were experienced tourists and many also cycled for commuting and/or errands. Average age was 41 years, with 9 years of frequent cycling. Their preferences, on a scale of 1-5, are given in Table 1.

Table 1: Preferences of Touring Cyclists for Commuting Conditions
 Preferences Score
Bike Lane 4.1
Wide Curb Lane 3.8
Bike Path 3.1
Trail 2.0
Sidewalk 1.9
Dirt Road 1.7
Safety 4.2
Quick Route 3.9
Direct Route 3.8
Smooth Pavement 3.8
Low Traffic Volume 3.6
Slow Traffic 3.3
Convenience for Errands 3.2
Avoid Hills 2.2
Scenery 2.0
Motorist Education 4.4
Youth Bike Safety Education 4.1
Adult Bike Safety Education 4.0

With increasing age and experience, these cyclists tended to greater preference for bike lanes and wide curb lanes and lesser preference for paths, trails, sidewalks, and dirt. With increasing age and experience, these cyclists tended to give greater importance to directness of route and road surface quality, and to give less importance to safety, traffic volume and traffic speed. With increasing age and experience, these cyclists give increased importance to all three educational programs.

2.2 Evaluation

This study confirms what many of us have already noticed. With increased experience, cyclists tend to shift their preferences toward vehicular-style cycling, to reduce their concerns about safety in general, and to increase their realization that education, of both cyclists and motorists, is the more important measure for attaining safety.

3 Sorton, Alex, and Thomas Walsh: Bicycle Stress Level as a Tool To Evaluate Urban and Suburban Bicycle Compatibility; Transportation Research Record 1438

3.1 Description

This is a simple method of rating roadways for cycling compatibility that calibrates the more simple features of earlier mathematical methods by viewing cyclist-eye videotapes of roadways. Three primary variables are used: curb lane traffic volume, speed of motor vehicles, and curb lane width (other variables might later be added: commercial driveway frequency, parking turnover, percentage of heavy vehicles). Each variable is rated on a scale of 1 to 5. Curb lane traffic volume is rated linearly between 50 and 450 vehicles per hour. Curb lane width is rated linearly inverse between 10 and 15 feet (3.3 - 4.6 m). Motor vehicle speed is rated approximately linearly between 25 and 45 mph (40 - 75 kph). The range for each variable was selected according, approximately, to the minimum and maximum experienced in practice. The final score is the average of the three.

Obviously the final score can range between 1 and 5 inclusive. The score is related to characteristics of cyclists according to the following schedule.

1: Reasonably safe for all types of cyclists. (The study specifically excludes children of elementary-school ages.)

2: Can accommodate experienced and casual cyclists, may need altering to fit youth cyclists.

3: Experienced cyclists, casual cyclists if "compensating condition" is present.

4: Needs altering or compensating conditions even for experienced cyclists, not recommended for casual or youth cyclists.

5: May not be suitable for cycling.

The authors tested the system by showing video recordings of known conditions to groups of cyclists who were individually classified as experienced, casual, or youth. Each cyclist was asked to rate one factor at each viewing, giving 1 for being very comfortable and 5 for not wanting to ride with this condition under any circumstances. The scores given by all the cyclists of each class were combined and the linear regression calculated and compared to that given by the assumed rating system. For all three factors, the test cyclists rated the change in stress as the factor varied as considerably less than was used in the initial system. As one would expect, the experienced cyclists accepted more intense conditions for equal stress than did either the casual or the youth cyclists. The R-squared correlation values ran from 0.94 to 0.13, with strongest correlation for traffic volume, somewhat less for vehicle speed, and very poor for lane width.

3.2 Evaluation


3.2.1 Improper relationships between variables

The suitability classification system does not properly consider the relationships between accident rate, cyclist skills, type of cyclist, and stress. Accident rate versus specified conditions

The suitability classification system starts by saying that level 1 conditions are "reasonably safe for all types of cyclists," and progresses to level 5 conditions which are described as "may not be suitable for bicycle use." The implication is that the accident rate increases as the severity of the specified conditions increases, becoming unacceptably high for condition 5. However, we know that only a very small portion of accidents to cyclists are caused by motor traffic from behind, which is the only condition considered. There is no known relationship that allows us to predict total accident rate from merely the three factors that this study considers. Required skills versus specified conditions

The suitability classification system specifically states that there is a relationship between the specified conditions and the skills that are required of cyclists. It states that "the on-street bicycling skill level of experienced bicyclists allows them to use the most direct and convenient routes, which often are the arterial or collector streets." This clearly distinguishes two kinds of characteristics. It says that arterial and collector streets require particular skills that are not needed on residential streets, and it says that experienced cyclists possess these skills.

However, the study does not specify those skills; all it considers are the conditions of motor traffic volume and speed and outside lane width. I know of no study that shows that increases in the intensity of these factors, singly or together, demand new skills of cyclists. Rather than that, we know that safe cycling requires that the cyclist exercise all of the five basic traffic skills, and that multi-lane roads require that the cyclist be somewhat better than is required on two-lane roads.

Instead of using that knowledge, the study rates roads according to conditions that have no reasonable connection with level of skill required. The only skill that is required to stay out of the way of motor traffic from behind is that of riding in a straight line near the right-hand edge of the roadway. That is the simplest traffic skill that exists and even elementary-school children, who are so unskilled that they this study excludes them, should be able to practice it. Type of cyclist versus specified conditions

The study specifically classifies cyclists into secondary-school, casual (recreation, utility, shopping, etc.) and experienced (commuting, touring and recreation). This sounds as though it is a classification by trip purpose. That is, it distinguishes shopping trips from work or school commuting trips, or at least it distinguishes the cyclists who make such trips. I know of no study showing that these types of trips require different skills, or take place under different circumstances, or are taken by cyclists with different skills or levels of experience. Therefore, the classification of trip purpose by skill level is inaccurate. The explicit inclusion of skill levels does not fit the facts, as described above.

However, in addition to the level of skill supposedly attained by experienced cyclists, the study describes preferences, saying that "casual cyclists tend to give high priority to avoiding congested, heavily trafficked streets." If this is a study in preferences, then it should say so explicitly. If that is what it is, then all that it demonstrates is that those cyclists who don't like to use heavily trafficked streets do not like traffic.

3.2.2 Are Different Accident Rates Acceptable for Different Cyclists?

The only accident types that vary according to the motor traffic volume, speed, and width of outside lane are those in which the cyclist is hit from behind. Consider the difference between a high volume curb lane and a low volume curb lane. Presumably, the rate of accidents of this type varies according to the traffic volume. That is the assumption of this kind of study, and it is the most reasonable assumption given the absence of better data. The only action that the cyclist can take to avoid this kind of accident is to ride in a straight line; so long as the cyclist does not swerve in front of overtaking traffic, this kind of accident is purely the fault of the motorist. As is pointed out elsewhere, the skill of riding in a straight line is the most elementary traffic skill of all.

However, the study explicitly says that high-volume streets are acceptable for use by experienced cyclists but not for casual or secondary-school cyclists. This is equivalent to saying that higher accident rates are acceptable for experienced cyclists than are acceptable for casual cyclists. What is the justification for such a distinction? Why should society accept higher accident rates for experienced cyclists than for inexperienced cyclists? That is the opposite of what we find in practically all other fields, where society reluctantly accepts a higher accident rate for beginners than for experienced operators.

If the rating system is intended to be a surrogate for accident rate, then it needs to consider the causes of accidents and the system that is devised to do this must be calibrated against valid accident data. Sorton and Thomas have attempted neither of these activities. Therefore, all statements about safety or suitability that reflect or imply safety must be removed.

3.2.3 Difference Between Initial Estimates and Ratings By Cyclists

Perhaps the rating system is not intended to rate either safety or skill. After all, Sorton calls this a stress-rating system. So what is it measuring?

Or what would it be measuring if it was recalibrated in the light of the difference between the evaluations of the cyclists and the initial estimates? The most important finding is that all the ratings given by the viewers attributed much less increase in stress as the conditions intensified than was predicted by the initial estimates. The difference is least according to volume. Where the initial estimate was a total possible change in stress produced by change in volume of 4.5 units, the rated change was 3 units. For motor traffic speed, where the total estimated change in stress was 4 units, the rated change was only 1.5 units. For lane width, for the estimated total change in stress of 4 units, the rated change was 1 unit.

Sorton and Walsh write that the initial estimates were "extrapolated from transportation engineering literature covering motor vehicles and then related to the bicycle stress level process. (The logic behind this extrapolation is that if there are problems for motor vehicles, these will be bigger problems for bicycles.)"

This is not valid logic. The operation of overtaking motor traffic has little effect on the operation of bicycles until the cyclist needs to change lanes, and then the primary consideration is length of gap available, which for streets with heavy traffic is determined more by the location and timing of traffic signals than by the traffic volume itself. The lane width has some effect, because narrow lanes require the cyclist to ride a straighter line with less margin, but that effect occurs regardless of traffic volume.

The actual initial estimates were made using the typical range of actual values. That is, maximum lane volume achievable on conventional streets ranges between 450 and 800 vehicles per hour, depending on traffic signal operation. Therefore, Sorton set 450 vphpl as stress level 5. Similarly, there are typical ranges for outside lane widths and for motor vehicle speeds. Sorton set the most intense condition as stress level 5.

The ratings made by the cyclists viewing the videos showed that they attributed much less stress to changes in the conditions than Sorton's initial estimates. Surely, this indicates that these factors are not all important and that other factors must be considered.

3.2.4 Measuring Emotions

Even if the rating system is accepted at its face value, what is being measured? Sorton and Walsh understand that they are measuring a subjective emotion, because they suggest further research to determine whether cyclists from other cities than Madison, Wisconsin, give the same or different ratings for the same videos that were made and viewed in Madison. They suggest that "it may be that bicycle stress level depends on the population size of urbanized areas."

Emotional data will change with circumstances. Sorton's data demonstrate that the effect varies with experience. Therefore it probably does change with location, because different locations provide different experiences. We can presume that it is likely to change with time and even with political circumstances. This is not so far-fetched, since the emotion that is being measured is fear of motor traffic from behind, a fear that has been largely created by the motoring establishment and which is embodied in the measuring system itself. We know that this fear has very little bearing on the cyclist accident rate or on the skills required of cyclists. Why then should we consider it at all?

3.3 Conclusions

There is a great difference between studying a phenomenon and making recommendations for policy. All knowledge is valuable and it is not appropriate to question the motives of those who seek it. However, when that knowledge is bound up with recommendations, the motives of those who make them and the results of their recommendations should be assessed. The Sorton-Walsh study calls itself a method for assessing the cycling suitability of roads rather than a study of the emotions of different groups of the public; it must be evaluated with its intended purpose in mind.

3.3.1 Regarding Propensity to Cycle

One reason why we should study the fear of overtaking motor traffic is that it may be an important factor in determining whether a person takes up cycling transportation. If public policy holds that cycling transportation should be encouraged, then we need to consider ways to attract people to using it. If fear of overtaking traffic prevents people from starting to use cycling transportation, then we should consider what to do about this fear.

Sorton and Walsh's study presents one solution to the problem: falsely assume and falsely state that the fear reflects actual danger and reflects the level of skill needed to prevent accidents, and rate streets according to the degree of fear that inexperienced people feel about them. That has the advantage of not having to change people's opinions, but it has the disadvantage of perpetuating dangerous and harmful falsehoods.

The falsehood is the supersitition that the dangers of cycling are those of motor traffic from behind. This fear of overtaking traffic tells people not to ride at all because they are using the roads that belong to cars. It tells those who decide to ride despite the dangers of which they have been told that their prime need when cycling is to stay out of the way of the cars that are using the roads. This is the cyclist inferiority phobia. It is absurd to expect that a population that believes that these superstitions are physical laws will develop much cycling transportation when it has motoring available to it, and it is very likely that many of those in such a population who do choose to ride will do so incompetently. The resulting cycling transportation will be biased to short distance, slow speed, and inefficient cycling with low social standing.

This is of course the perpetuation of the present American system, in which there is very little cycling transportation, which is done by only a very small portion of the population, while those few who perform the bulk of it are those who have largely managed to overcome the general public superstition about cycling in traffic.

4 Sorton, Alex; Measuring the Bicyclist Stress Level of Streets; ASCE HPT Committee, November 1995

This is a follow-on to Sorton's initial paper about measuring the stress felt by cyclists in Madison, WI, when observing videos of various streets. In this paper, Sorton compares the responses of a group of Chicago Bicycle Federation cyclists to the same videos that had been seen in Madison. Chicago cyclists rated the streets as about one-unit (on a scale of 1-5) less stressful than did the Madison cyclists.

5 Davis, W. Jeffrey; Bicycle Test Route Evaluation for Urban Road Conditions; ASCE HPT Committee, November 1995

5.1 Description

This is the second, simpler form ot Davis's mathematical function for rating the cycling suitability of roads. His earlier one included specific terms for evaluating intersections, and is described in Epperson's paper that is described below. The function used here gives the cycling suitability rating as the sum of a series of terms for various conditions: motor traffic volume in the outside lane, motor traffic speed, outside lane width, the condition of the pavement, and the condition of the location. In this system the higher the rating the worse the suitability for cycling. The maximum possible value and the approximate average value in this study for each term are given in Table 2.

Table 2: Rated Characteristics & Ranges
 Characteristic Max ~Avg
Lane Traffic Volume 4.0 1.4
Motor Speed 1.0 0.6
Lane Width 2.5 1.5
Pavement Surface 4.0 ?
Location Conditions 4.25 ?

One can see that although each term is given equal weight in the formula, the range of values of each term does not give equal weight for the characteristic in the total score. If we look at the range from average to maximum as a reasonable measure of the relative effectiveness of that characteristic in influencing the score, we see that traffic volume has a range of 2.6, lane width a range of 1.0, and motor speed a range of 0.4. These are the effective weights given to each of these factors. Davis thought that he gave equal weight to each factor, but this analysis shows that a different interpretation is more reasonable.

The individual factors that contribute to pavement smoothness and to locational conditions also should be questioned, both for the rating of each and of the factors that are not included.

The factors for pavement smoothness are: cracking, patching, weathering, potholes, rough edge, curb and gutter, rough RR crosssing, and drainage grates. Certainly all of these things, if properly defined, detract from cycling suitability, but how many points off for how many defects in what length of road is a great question. Among many satisfactory drain grates one can find one that is very dangerous; do we count the average or the worst, or how does one rate the roadway? If the factor value for patching is 0.25 (as is given), does one patch in a mile rate as 0.25, or as 0.01? Does it make a difference if the patch is satisfactory or not? These are not explained. A pavement factor that is left out is concrete vs asphalt; concrete often has bumpy expansion joints and sometimes has dangerous ones parallel to travel.

The adverse factors for location conditions are: angle parking, parallel parking, right turn lanes, severe grades, moderate grades, frequent curves, short sight distance, numerous driveways, industrial land use, and commercial land use. Favorable factors (producing a negative contribution to the score) are: physical median, center turn lane, and paved shoulder. In contrast to the listing of right-turn-only lanes as a detriment to cycling, they are a benefit for cyclists, because they move the right-turning motorists to the cyclist's right before the motorists turn right. That avoids the motorist-right-turn car-bike collision, which is about number 2 or 3 in the frequency of motorist-caused car-bike collisions. There is no listing for the beneficial presence of traffic signals with protected left turn phases, which eliminate, where installed, the motorist-left-turn car-bike collision, which is the most frequent of motorist-caused car-bike collisions. As an additional benefit, they also make left turns easier for cyclists. As for listing paved shoulder here, if it is suitable for cycling (smoothness, cleanliness, etc), then it should be listed as part of outside lane width, while if it is not suitable for cycling it shouldn't be listed at all. are frequent curves detrimental to cycling? Some of the nicest roads I know have frequent curves. Furthermore, there is no consideration of the features that are very important because they speed or delay cyclists. That is, a street that is protected from cross traffic is much safer and faster than one that is not. I have partaken in a discussion with people who maintain, as probably most of the subjects in this study do, that residential streets are much safer than arterial streets. That may be true for slow cyclists, but it is not true for fast cyclists, who should slow down for every intersection if they are not to take dangerous and unlawful risks. Being protected by stop signs is generally ideal, but no street is so protected for all of its length. Practically all streets that are protected by stop signs eventually come to intersections with unfavorable stop signs or traffic signals, and generally those traffic signals have an unfavorable green proportion because the traffic on the cross street is more numerous or faster. Therefore, the system should consider the disbenefit of intersections in general, with the value of being protected by stop signs and the disbenefits of facing unfavorable stop signs, and have a system for evaluating the effect of traffic signals.

Davis then surveyed a set of road segments that formed an easily cycled route, and scored them according to his system. He then had a sample of cyclists ride this route and report their scores for eleven different factors concerned with roadway and traffic. (See List 1) The scores were all rating on a 1 to 10 scale, not in any way a reflection of actual values. He then analyzed the cyclists' scores for each factor and compared them to the ratings he produced by using his system.

List 1: Rated Characteristics

1: Pavement surface
2: Lane/intersection configuration
3: Outside lane width
4: Curves, sight distance, visibility
5: Hills, terrain
6: Adjacent land use
7: Traffic volume
8: Motor vehicle speed
9: Parked vehicles
10: Driveway traffic
11: Traffic signals, signs

5.2 Analysis of Results

The first thing that Davis recognized, probably in advance, was that the subjects could not rate length of trip and time for trip, as they were not going to any particular destination and had no particular schedule to meet. He covered this by asking his subjects to rank the relative importance of the four factors: Length of route, travel time, roadway conditions, and traffic conditions. Starting with the least important, their ratings were travel time, length of route, roadway conditions, and traffic conditions. Therefore, concluded Davis, such a rating system, if validated by the responses of subjects on test rides, would reflect two factors most important to cyclists, traffic conditions and roadway conditions.

I think that this is dubious. Commuting cyclists whom I know tend to evaluate routes according to two criteria: the one that takes the least time, for when time is important, and the one that provides the most physical conditioning within the available time, for when that is desired. I think that rating routes for cycling suitability must consider at least their transportational utility. That may be combined with a rating of the road's conditions according to some system such as Davis's, but it should not be ignored.

Davis provides comparisons of three ratings: the calculated rating, the overall general rating given by the cyclists at first, and the average of the ratings given by the cyclists for each category. The calculated ratings varied most, the general ratings varied less, and the averaged ratings varied very little. One would expect the averaged ratings to vary less than any individual ratings, as actually happened. However, the cyclist's general ratings were not a component of the average. This shows that the cyclists did not accurately average the ratings that they later wrote down and went more on general impressions. The general ratings and the averaged ratings varied consistently together, while the calculated ratings in 3 out of the 8 instances varied in the opposite direction from the other two. This shows that the cyclists used a different system than the formula originally hypothesized.

5.3 Evaluation

Davis did not provide a comparison between the actual values and the cyclists' evaluations. Indeed, the two systems were incompatible. However, it is possible to compare the cyclists' ratings against the three numerically defined characteristics: traffic volume, traffic speed, and lane width. I compared the average of the cyclists' scores against the given conditions; the correlation coefficients, assuming linear relationships, are given in Table 3.

Table 3: Correlation Coefficients
Characteristic  r2
Traffic Volume 0.90
Traffic Speed 0.80
Lane Width 0.003

These show that the cyclists' satisfactions were strongly related to traffic volume and traffic speed, but that cyclists had no consistent way of judging whether the roadway width was or was not adequate. Note that the order of the correlation coefficients is the same as in Sorton's study, and the values are also similar. Each cyclist may have known how wide the outside lane actually was, but if so the cyclists disagreed about how satisfactory it was. Consider the facts. The narrowest road, with 8 foot lanes, carried only 750 cars per lane per day at 20 mph, while the next narrowest lane, 10 feet, carried, in one example, 5,000 cars per lane per day at 40 mph. With variations like that in the actual conditions, it is no wonder that the cyclists could make no consistent relationship between satisfaction and lane width. Davis recognized the deficiency, but put no numbers to it.

This study tells us that average cyclists pay fairly accurate attention to the volume and speed of same direction motor traffic in assessing their estimate of the suitability of cycling routes. It implies, but does not prove, that they pay much less attention to anything else. However, more can be shown than Davis did. If the average cyclist scores of satisfaction with motor-vehicle speeds and volumes are plotted on the same plot as Davis's Figure 3, Rating Index Comparison, the following information appears. In five out of the eight routes, the level of satisfaction with speed is identical to that with volume, and in the remaining three instances the values are only one unit apart. Furthermore, if the average of the two is plotted its swings are always in the same direction as the average rating and the general rating, but are more extreme. In other words, all the other factors, in total, serve to dilute the effect of motor-vehicle traffic on the opinions of average cyclists.

However, while the opinions of average cyclists express reality in the sense of being a fairly accurate measure of same-direction motor traffic, there is no indication at all that they are attached to reality in any substantive way. Cyclists pay great attention to same-direction motor traffic, but what is the actual effect of same-direction motor traffic upon cyclists? The effect is very small. The most significant effect is that it makes them allow more time when preparing to turn left, because adequate gaps are less frequent as the speed and volume of overtaking traffic increase. There is also an increase in the frequency with which cyclists encounter right-turning motorists., which is presumably proportional to the number of motorists who overtake the cyclist. However, I venture to suggest that these considerations play no part in the opinion of average cyclists, and they certainly were not taken into consideration in the study.

What then were the cyclists considering? Davis did not ask. I venture that the cyclists considered the high volume roads dangerous, and the only danger that they were considering was the probability of being hit from behind. That probably is proportional to the number of overtaking motorists, but it is only a minute cause of casualties to cyclists under urban, daylight conditions.

In short, there is no real reason to consider that the speed and volume of same-direction motor traffic is a real measure of the cycling suitability of most routes. However, there is great reason to consider the effect of this cyclist-inferiority superstition on the type of planning and of programs that we conduct for cycling transportation, and to consider how to produce a rational program in the light of this present false superstition. So far I see every indication that government is catering to the superstition and doing nothing at all about producing a rational program.

6 Epperson, Bruce; Evaluating Suitability of Roadways for Bicycle Use: Toward a Cycling Level-of-Service Standard; Transportation Research Record 1438

6.1 Description

Level of service (LOS) originally referred to the ability of drivers to achieve the speed that they preferred in a flow of traffic. Level A meant free travel, level F meant traffic jam. Level of service can be either observed or estimated by complex calculation taking many factors into account. LOS for transit riders and pedestrians is measured in space available.Epperson dismisses the idea that either concept of level of service would be significant for cyclists.

Instead, Epperson advances the concept that level of service for cyclists should be a measure of the probability of having an accident while traveling the route. He gives a detailed account of Davis's initial formula, which included an evaluation of intersections as well as of the road conditions between intersections (intersections were not evaluated in Davis's experiment of 1995). Epperson discusses some problems with Davis's rating system that give irrational results under some circumstances, but concludes that Davis had identified "the three critical factors that affect the comfort, convenience, and perception of safety common to virtually all bicycle users: per-lane traffic volume, traffic speed, and lane width." Epperson emphasizes his view by writing that all the other factors that Davis considered merely "tended to dilute the focus of the evaluation model on the three critical factors of roadway speed, per-lane volume, and lane width."

However critical these factors may be, Epperson acknowledges that attempts to correlate Davis-style ratings with accident frequency or probability have failed. Of course, without adequate counts of bicycle traffic volume, which had not been obtained in any test, no safety-rating system can be properly tested.

Then Epperson describes the use of such rating systems to fulfill political demands. In areas with poor automobile LOS, Florida law demands remedial measures before granting any further development permits. One of the remedial measures that can be applied is to provide roads that are good for transit and cyclists, as demonstrated by having low LOS ratings according to some system similar to the Davis system.

Epperson describes the use of the Sprinkle Consulting LOS formula developed by Landis (see next review). When using this formula, Epperson states that a LOS of "4.0 or lower was determined to provide an adequate level of service for less experienced cyclists or children. A rating of 5.0 or lower was judged to provide an adequate level of service for more experienced cyclists."

Epperson then considers the meaning of LOS for cyclists. He gives his analysis of the schism between "casual and experienced" cyclists, referring to my book Bicycle Transportation . His description says that all types of cyclists like wide lanes, low traffic volumes and low traffic speeds, but that experienced cyclists are more willing to trade off those characteristics for the higher speeds and shorter travel times available on streets with heavier traffic. He concludes that bicycle LOS ratings are a vital part of transportation planning because they reflect how the general population will use the road system by bicycle.

6.2 Evaluation

As a first criticism, Epperson's dismissal of the traditional concept of level of service as inapplicable to cyclists is incorrect. That concept is eminently applicable to bike paths and university campuses, where the problems of differing speeds and differing patterns of use produce very low levels of service on well-used paths. While the theoretical capacity of bike paths is very high, the level of service falls drastically at volumes that are only a small fraction of the maximum that is attainable. The productivity of bike paths, measured as bike-miles per hour per foot of width, is never near the theoretical values calculated from the area that the bicycle requires and the speed at which it can travel. Epperson's conclusion on this subject merely means that he predicts that there will never be sufficient bicycle traffic on any road for the cyclists traveling in the same direction to significantly interfere with each other.

The rest of Epperson's analysis asserts that the amount and distribution of bicycle traffic will be determined by the distribution of same-direction motor traffic and will be strongly directed away from high levels of that traffic in accordance with present experience.

There is only one scientific justification for such a conclusion. That is, that present conditions will continue. That predicts that future cyclists will be as poorly trained, as inexperienced, and as frightened as present cyclists are. If cycling transportation is as poorly used as it is at present, then that prediction is reasonable, but continuation of the present low volume of cycling transportation is no the intent of any transportation program that considers cycling at all. The objective of such programs is to develop cycling into a significant part of urban transportation. If that occurs, it must change the type of cyclist. It is impossible to have a successful cycling transportation system without developing experienced cyclists.

That does not necessarily imply that all experienced cyclists act the same; after all, Holland and Denmark have experienced cyclists who act like our inexperienced ones. However, they operate in an entirely different physical environment that we will not adopt, that we could not adopt in less than many decades. American urban cyclists have to ride much greater distances than European ones, which requires greater speed over the most direct routes. It is more reasonable to predict that American urban cyclists operating in a future successful urban cycling transportation system will continue to act like those few who do so today, than that they will change into the slow, short-distance cyclists typical of the European systems. This presumption that American urban cyclists will continue to be casual cyclists and children is the most devastating of Epperson's errors.

Epperson is led into this error by his misunderstanding of what he calls the difference between casual and experienced cyclists. As he states it, "the so-called schism is not a debate about the virtue of [wide lanes, low traffic volume, and low traffic speed] but is instead a different propensity for various types of cyclists to trade off a pleasant riding environment" for greater transportation utility. The point is not one of losing a pleasant riding experience. If that were all, there would be no schism, which is presumably why Epperson describes it as a so-called schism.

What Epperson does not understand is that the inexperienced cyclist is frightened of heavy same-direction traffic, and that as urban cyclists gain experience, and even more as they gain understanding of the traffic system and the skill of operating in it, they largely lose that fear. They come to understand that most of their dangers are visible in front of them and that they have methods of dealing with those dangers. Sure, with everything else equal, they would prefer a low-traffic route, just as motorists would, but they no longer consider this the most important basis for choosing a route.

For these reasons, I consider Epperson's paper to be an explanation for the reasons why, and the route by which, bike planners have retreated from an analysis of the actual safety of different types of routes to a political stance that is based on current superstition and ignorance about cycling in traffic.

7 Landis, Bruce; Bicycle Interaction Hazard Score: A Theoretical Model; Transportation Research Record 1438

7.1 Description

Landis describes a more advanced model than the Davis or the Sprinkle Consultants models, which he calls the Bicycle Interaction Hazard Score (IHS), and proposes a method of validating it. However, while his model may, or may not, be one of hazard estimation, his proposed method of calibrating it is entirely a survey of superstitious opinions.

Landis starts out with the distinct improvement of considering the hazards of both same-direction traffic and of crossing traffic. His formula is:

IHS = {ADT/L*(14/W)^2*[a1*(S/30)*(1+HV)^2 + a2*PF] + a3*LU*CCF}*0.1


ADT = average daily traffic

L = number of through lanes

W= usable width of outside through lane

LU = land use intensity: residential single family =1. If any other use >30%, then = 15

CCF = curb cut or parking frequency, per mile

PF = pavement factor (reciprocal of FHWA PAVECON factor)

S = speed limit

HV = proportion of heavy vehicles

a1, a2, a3 = calibration coefficients

Landis states that these data items are easy to collect. On the basis of interviews with groups representing experienced, casual, and child cyclists, Landis decided that the following values were appropriate for the calibration constants:

a1 = 0.01, a2 = 0.01, a3 = 0.02. This means that the score is made up of the following contributions:

Table 4: Contributions to Hazard Index Score
 Characteristic Factor
Same-direction Motor Traffic 0.79
Pavement Condition 0.13
Driveway Traffic 0.08

Landis provides a table showing the changes in the total score produced by particular changes in each variable. He describes a system for obtaining the opinions of a greater number of cyclists over a large number of road segments and for determining the correlation between personal and video observations of the same roads to cut the cost of obtaining further opinions.

7.2 Evaluation

For all that Landis calls this a bicycle interaction hazard score, this is nothing of the kind. It is purely an opinion survey of the general public that is based on particular factors that are easy for planners to collect and are felt to be important by the general public. That is all that it is in its present form and that is all that Landis proposes it to ever be. While Landis initially hypothesized that the contribution of crossing traffic to his index, whatever that index might be, might be important, in his development of that index he eliminated all crossing traffic except parking traffic.

8 The studies as examples of bicycling program policies

8.1 The recommendations have no scientific justification

None of the authors advances any reasons for believing that their procedures provide an accurate measurement of anything. If they measure anything at all, they measure a fear in an abstract setting according to an intellectual kind of consideration. That is, it asks "How frightened would you feel if you were to cycle on a street with these conditions of overtaking traffic?" No psychologist would consider any of these studies a reasonable study of fear. I think that we need to study this fear, but we need to study it in a way that reflects psychological knowledge and technique. I think that we need to study this fear because it is a prime factor in our problems with cycling transportation. Once we know more about it, we may be able to devise better ways of dealing with it.

We need a sound policy for cycling transportation, and such a policy has to consider the fear that the general public has about cycling in traffic. However, the recommendations given in these papers for dealing with that fear have no scientific basis for being part of that policy. Following their recommendations would not only be scientifically dishonest, but it is poor policy because it gives official support to false fears about cycling.

8.2 The recommendations appeal to the general public superstition

There are several reasons why the average person finds these recommendations to be psychologically seductive. The most basic reason is that he incorrectly believes that the prime danger to cyclists is the motor traffic from behind. Since he believes this, the thought that bicycle planners may be wrong doesn't cross his mind. More than mere intellectual quiescence, any program that questions the basis for the biccyle planners' recommendations threatens the belief on which he relies for his own safety. Furthermore, since he knows that he cannot do anything to avoid the great danger of being hit from behind, he must consider that those who ride in traffic are either risk-taking fools or possess some super-human ability to bounce cars off their backs. This means that safe cycling in traffic is not a matter of skill and judgement but requires magic powers that are unavailable to ordinary mortals. For most Americans, these are psychologically compelling motives.

There are other reasons why the bicycle planners' recommendations are easy to believe. Their program promises safe cycling without the cyclist having to do anything or take any responsibility. The program promises motorists (and most American voters are motorists first and are very unlikely to be transportational cyclists at all) to get cyclists out of their way on the roads that they most want to use. The program does not threaten anyone with becoming a cyclist or even considering becoming one; if it persuades someone else to become a cyclist, that's all to the good, and it will keep him off the roads that I want to use.

There is also the pragmatic argument that it is often easier to change physical conditions than to change people's behavior. However, the conditions under which this principle applies are not present here. The changes in conditions (some designated bicycle routes, some widening of curb lanes) cannot compel people to cycle. All they can do is to persuade, but at the same time the policy that justifies these changes is based on, and strengthens, the false superstition that most of the roads are too dangerous for cycling and that individual cyclists cannot do anything to improve their chances. In short, a policy based on the bicycle planners' recommendations strengthens the cyclist-inferiority superstition, which is exactly the incorrect policy. Given these mutually opposed forces, any cycling transportation that is generated will be less than would naturally occur if cyclists understood the truth, and will tend to undercut and denigrate the position of those who, today, are doing the bulk of the cycling transportation.

In summary, there are many reasons why the bicycle planners' recommendations exert powerful psychological attraction, but none of those reasons is scientifically justifiable.

8.3 Specific criticisms

If the formulas that were used actually considered the traffic hazards of cycling, they would contain a very large component of crossing traffic because crossing or turning traffic causes about 95% of car-bike collisions. A distribution that matches the accident distribution could be constructed from the factors responsible for injuries and deaths, not the fault of the cyclist, estimated in Bicycle Transportation, Table 5-16. as calculated in Table 5:, Factors in Accidents.

Table 5: Factors in Accidents
 Factor % of all accidents % weighting
Road Surface Defect 20 58
Turning & Crossing Motor Traffic 8 23
Same-Direction Motor Traffic 0.4 1
Pedestrians 4 12
Dogs 2 6
Totals 34.4 100


Of course, each of these factor weightings would have to be allocated among the types of accidents covered according to frequency, and then the appropriate coefficient for the relationship between the condition that can be measured and the accident rate would have to be investigated and measured. An example of such an item of data would be the probability that a cyclist would be hit by a left-turning motor vehicle for the frequency with which oncoming vehicles turn left or, in a slightly different method, approach the turning point. That is a very large job that hasn't been done for motorists, let alone for cyclists. However, by merely considering the gross-level weightings that would be required we can see how deceptive are the formulas that have been used. By considering the amount of investigation that would be required to produce a reasonably accurate accident rate prediction, we can see that this is not likely to be accomplished at any time, and certainly will not be available within the planning horizon.

Consider also the type and direction of the misjudgments that the creators of these formulas have committed. Epperson describes the original Davis formula, which did consider intersections, as including many factors considered detrimental to cyclists. Rather than considering these factors detrimental to cyclists, I consider many of them beneficial to cyclists and some of no particular effect at all. Cyclists benefit from right-turn-only lanes, from left-turn-only lanes, and from traffic-actuated signals. Cyclists are not bothered by dual left-turn-only lanes, or by more than one through lane. All of these are considered by Davis and Epperson to be detrimental to cyclists. Furthermore, Davis's formula contained a term that gave far more importance to the volume of same-direction traffic than to the volume of crossing traffic. If the volume of crossing traffic is only 20% of that on the direct road, the term calculates to 1.67, while if the volume of crossing traffic is 500% of that on the direct road, the term comes to only 0.33. That is exactly the wrong relationship.

Another misjudgment is the failure to properly consider the fear factor. All of the creators of this system consider that the intensity of same-direction traffic is actually a major factor affecting cycling safety. They consider that its effect on cyclists is one of actual danger that requires particular skills to avoid. They do not consider that cyclists are frightened, but consider that they are merely exercising good judgement in the face of some actual danger. Nowhere do they consider what skills the cyclists must have to avoid this supposed danger from behind, skills whose presence in some cyclists would justify, in the minds of the planners, those cyclists' use of more dangerous streets. In actual fact, there are no such skills. Because there are no such skills, the danger cannot exist, or experienced cyclists would have very high accident rates, whereas they have the lowest accident rates.

In actual fact, of course, most of the dangers to the cyclists are ahead of them where they can see them and for which they have specific countermeasures. The change as cyclists acquire experience and knowledge is not one of accepting greater risk, but the realization that the danger is far smaller than they first feared, and largely comes from a different direction.

While it is entirely reasonable to measure the present opinions of today's people with little cycling experience, it is unsound to presume that the same opinions would continue in a society in which transportational cycling was more frequent. It is far more reasonable to consider that the opinions of future regular users of cycling transportation would more likely resemble those of the present expert users than those of the present non-users or casual users.

The reader should not forget the failure to consider the distances and speeds of cycling that typical American cities require, and their effect on the type of cyclist they develop, and the failure to consider the development of cyclists in a developing system of cycling transportation, as discussed above.

Mistakes such as these show that those who created the hazard index systems had a very inaccurate view of cycling traffic. The inaccuracy of that view probably contributed to the subsequent degradation of the estimating process that they committed.

8.4 General evaluation

The studies as a whole describe a series of changes that is better described as decay rather than as progress. Initial intents to develop a system of rating the hazards of particular journeys by bicycle, which presumably would have an ascertainable relationship to the accident pattern, all turned into gross surveys of the superstitious opinions of uninformed people with, in at least one instance, deliberate intent to steer the political process. Furthermore, the actual nature of the process was concealed behind deliberately misleading names and pseudo-scientific procedures that enable city planners to provide a specious justification for their plans.

8.5 The Responsibility of the Bicycling Committee of the Transportation Research Board

The Sorton and Walsh paper was published by the Committee on Bicycling and Bicycle Facilities of the Transportation Research Board, an arm of the National Academy of Sciences. Were I still reviewing papers for that committee, I would have recommended rejection for the reasons stated herein: there are too many obvious scientific errors.

Possibly, publication of the research data would be justified as an analysis of public opinion about cycling, provided that that was investigated in a manner suitable for such a subject. Such a publication would necessarily have to include an evaluation of the errors in such opinion.

The Bicycling Committee of the Transportation Research Board is supposed to be composed of those who are most competent in the nation regarding knowledge of cycling transportation. Unfortunately, the committee has never lived up to that supposition. The committee of fifteen years ago, when I knew it well, would have accepted this paper, just as the present committee has. Rather than basing its acceptances on scientific accuracy, the committee has too frequently judged according to the degree to which a paper agrees with the superstitions of the majority of its members and the corresponding policies of the governmental organizations in which many of them work.

9 Strategy for cyclists

9.1 Criticizing bike planning policies based on these formulas and concepts

Sorton and Walsh describe their purpose by writing that the cycling suitability of streets should be assessed when "developing a bicycle network to arrive at sound decisions on the appropriate locations of bicycle usage." The other authors have similar intents. In other words, either the public should be directed to use streets that have low ratings by the proposed standard or streets that provide desired routes should be changed to attain low ratings. In general, we have seen more attempts to direct cyclists than to change the road characteristics.

It is obvious that one result of such a policy is to direct cyclists away from the roads most used by motorists. Regardless of the effect on cyclists, that is a result that is desired by motorists. Therefore, the recommendation should be evaluated with that result in mind. If the recommendation makes little sense for cyclists, then it is reasonable to conclude even more strongly that pleasing motorists is a prime motive for the recommendation.

What benefit does the recommendation that diverts cyclists from the most obvious roads provide for cyclists? Sorton and Walsh do not provide any, saying only that "a procedure is needed that satisfactorily explains the effects of traffic volume, speed, and curb lane width on the different types of bicyclist." Antonakos simply collected information about the preferences of cyclists. Davis, Epperson, and Landis all started with the intention of selecting roads that are safer for cyclists, but all ended by merely measuring the fear of a largely imaginary danger.

Hence we must look elsewhere for the supposed benefit. The typical argument is that cyclists do not know the best routes to take and need to be directed to them. However, we do not do this for motorists, except in a very general way. For motorists we provide destination signs and through route number signs, which assist strangers in the area to reach particular destinations or to travel through the area and, to some extent, we hope that these will channel through traffic away from residential streets. However, we do not expect that local motorists with local destinations will be directed by these route signs. We expect these persons to discover and to use, without other direction, the best routes for their particular purposes, whether or not these are the signed routes. There is no obvious reason why cyclists require a different service; indeed, since few cyclists travel very far, it is more likely that a cyclist will have a good knowledge of the streets within his smaller range of activity than does a motorist with his larger range. Therefore, this argument does not explain either the need or the desire.

It is argued that few of the people who might cycle actually know much about the streets that they might cycle on. Therefore, so the argument goes, they need to be told. That is no answer; it merely raises the question of why we need to tell them. These people are not children who are being let out into the world for the first time; they have traveled around their own areas by walking and driving, at least as passengers and probably by being drivers. All they have to do is to get out and try the different routes that they already know, and decide which they prefer. The whole argument that people need to be told what their streets are like is utterly foolish, but many people make it and public policy is based on it.

Furthermore, the two characteristics that are considered most significant in most people's preferences are traffic volume and speed, and the tests that have been done show that people are quite good at estimating these by a few observations. It does not take an esoteric formula to tell most people whether they would feel safe on a particular road. That is not to say that their feeling of safety or danger has any relationship to actual danger.

When the argument is investigated, its base is the superstition that roads are so generally dangerous for cyclists and riding in traffic requires such an elite level of skill, that people who prospect for safe routes are too likely to be killed or injured before they find one. That argument is false in fact and the recommended method of designating safe routes according to the conditions of overtaking traffic is a false recommendation for a false policy. That is, the roads are not generally particularly dangerous for cyclists and the dangers that do exist are not defined by the conditions of overtaking motor traffic.

Sometimes this argument is redefined to say that even though roads with heavy traffic are not particularly dangerous, it is sufficient that those who do not cycle believe that they are. In other words, this version of the argument goes beyond the tacit, unquestioned, but false, assumptions of the method as used by Sorton and Walsh, and recommends lying to the public with malice aforethought to achieve some objective other than the good of the cycling public. That is implicit in the admission of the other planners that their systems measure preference rather than danger.

9.2 A proper policy for cycling transportation

In my opinion, honesty would be a much better policy. That is, to teach people what is actually known about cycling. We know the skills that are required for cycling in traffic and we have sufficiently accurate knowledge about the dangers of cycling and how to handle many of them. The skills and conditions are generally those that occur when cyclists act and are treated as drivers of vehicles. We know how to teach these lawful skills to those who want to learn them. We have generally adequate traffic laws that require other road users to operate properly with regard to cyclists. These conditions are sufficiently well met throughout the nation that those who choose to cycle properly can generally do so. We do have problems with the road system, but they are generally local and are not systematic, and they would be much more easily solved if the public accepted that cyclists are drivers of vehicles.

We need to face and overcome our problem that the public has been misled about cycling for decades. Cyclists are not inferior to motorists, they are not at great danger when they cycle properly, and they fare best when they act and are treated as drivers of vehicles. To correct this psychological untruth will be harder than making easy, but false, promises that changes to facilities will make cycling safe. However, telling the truth and acting on it is the only way that we can reasonably persuade more people to perform cycling transportation in distributed cities such as we have today.

Certainly, we need a systematic way for parents and guardians to limit the locations where their children may cycle, but that needs to be done on the basis of the skills of each individual child rather than on factors, such as the bike planners' three, that have little bearing on the ability of the child to operate properly in reasonable safety.

It is a truism to say that over the long run, truth is a far better guide than falsehood. With truthful knowledge about cycling, we can expect a reasonable amount of cycling transportation done in the best ways. That is the only kind of cycling transportation that can compete against the automobile in today's distributed cities.

Cyclists need to recognize that these city planning programs are based on thoroughly defective models designed by planners with a defective knowledge of cycling transportation and justified by appealing to the opinions of a public with, on average, no better information.

9.3 A policy for dealing with inadequate roadways

That is not to say that all measures to accommodate same-direction traffic are unnecessary or undesirable. Wherever the road carries excessive traffic, consideration must be given to handling that traffic. One method is reducing the amount of traffic or the space required by that traffic. This is the realm of large-scale city planning. One aim of large-scale city planning is reducing people's need for travel. Whatever may be done to achieve that objective, it will take a very long time and travelers require quicker solutions. Another method is persuading people to use mass transit, which requires less space per passenger. This works only in some locations, it also requires a long time, and so far has not had sufficient effect to improve cycling conditions.

Shorter-term measures are required, of which there are two. So far as cycle traffic is concerned, there are two problems: motorists get delayed and motorists squeeze by cyclists with insufficient clearance. While the first may seem more important to motorists and the second may seem more important to cyclists, in truth both parties dislike the situation. Cyclists do not like delaying motorists, and motorists do not like squeezing by cyclists, although they often prefer doing so to getting delayed.

If the outside lane has adequate width, these problems disappear. That is the best solution. If society chooses, for whatever reason, not to supply adequate width in the outside through lane, then we must rely on the existing legal principle and statute law that overtaking drivers may overtake only when they can do so safely, including giving adequate clearance. That means that where lanes are too narrow for safe overtaking within the lane, then motorists must change lanes to overtake. Some cities have posted signs saying just this in the locations where the problem is acute.

This also means that in these locations the speed limit, at least at times of heavy traffic, must be sufficiently low that motorists seeing a cyclist ahead have a reasonable chance to change lanes. A road that had two 12-foot traffic lanes and an 8-foot parking lane in one direction cannot be restriped to three 10.7-foot lanes and operated at 45 mph, because 10.7 feet is inadequate for safe overtaking of cyclists and 45 mph is too fast for the motorists approaching a cyclist to easily change lanes after seeing the cyclist.

These measures are adequate, so far as the problem of cyclists and excessive traffic is concerned. (Improvements in other areas are also necessary.) Anything further is catering to excessive fear of largely imaginary dangers. Many people argue that catering to present fears is an easier way to attract people to cycling transportation than is training cyclists. In the short run, that is correct. However, doing so has very serious disadvantages to both present cyclists and to the prospects for cycling transportation. The measures that cater to the exaggerated fear have two effects: they both endanger cyclists by paying attention to the lesser dangers while increasing the greater dangers, and society's endorsement of such measures justifies and strengthens the irrational fear. Noland, who shows few signs of being either sympathetic to cyclists or well informed about cycling transportation, makes the same argument that I have been making for twenty-five years: if we attract cyclists to particular facilities under claims of safety that turn out to be false, we then have a larger cycling population operating with a higher individual accident rate.

The problem with the city planners who have devised these road rating systems is that they believe they are considering the most significant accident problem and they believe that it is associated with facility design, while in reality they are measuring an exaggerated and completely misdirected emotional problem that results in misdirected behavior with dangerous consequences.


Return to: John Forester's Home Page                                          Up: Facilities