Skip to contentUnited States Department of Transportation - Federal Highway Administration FHWA Home
Research Home
Report
This report is an archived publication and may contain dated technical, contact, and link information
Publication Number: FHWA-RD-03-037
Date: May 2005

Validation of Accident Models for Intersections

FHWA Contact: John Doremi,
HRDI-10, (202) 493-3052, John.doremi@dot.gov

PDF Version (1.61 MB)

PDF files can be viewed with the Acrobat® Reader®

3. RECALIBRATION

This chapter presents recalibration results for the five types of rural intersections that were the subject of the validation exercise undertaken in the first part of the project. The first section provides a discussion of the recalibration approach. In the second section, the data and related issues are discussed. Third, AADT model estimation results are presented, followed by fully parameterized model estimation results. Sensitivity analysis results for the AMFs derived in this research then are given. Finally, a discussion and conclusions as a result of model recalibration are provided.

3.1 RESEARCH APPROACH

This model recalibration effort complemented the comprehensive model validation previously conducted as part of a larger technical evaluation of crash prediction models. It should be acknowledged that several anticipated end-uses of the crash prediction models guided all decisions made throughout this careful evaluation, which resulted in some specific overriding considerations while conducting the model recalibration:

  • The most likely end-use of the crash prediction models is embedded code within the IHSDM, with the sole intent to predict future crashes at intersections throughout the United States.
  • The models need to be able to predict the change in safety as a result of changes in traffic and geometric features relative to nominal conditions, corrected for intersection type and State- or regional-specific effects.
  • Environmental effects on safety, such as adverse weather and lighting conditions, while important factors, will be accounted for in State or regional correction factors.

Considering the likely end uses of the crash prediction models within the IHSDM, considerable time was spent identifying a strategy for recalibrating statistical models. A strategy was needed for several reasons. First, there were multiple levels and types of models in the source documents-requiring a prioritization of models to be calibrated. Second, there are numerous methodological approaches reflected in the source documents, which need to be prioritized. Finally, the treatment of explanatory variables is dependent upon the methodological approach taken. Before describing the research technical strategy, some guiding philosophical principles used to guide the model recalibration effort are presented.

It was felt that the majority of effort in the recalibration should be devoted to refinements to existing models. This includes changes to parameter estimates, and perhaps minor changes to model functional forms. This approach is based on the collective opinion that prior work, including the estimation of statistical models, was done carefully by experts in the field of transportation safety, and decisions such as variable selection, model functional form, and statistical model selection represent state-of-the-art knowledge with respect to intersection crash prediction models. Past documentation, critical evaluation, and discussion with other experts in the field confirm prior beliefs that the existing set of models represents a defensible and sound starting point. It is believed that moderate to serious departures from existing models should be accompanied by detailed and defensible descriptions of the how, why, and in what cases departures from previous methods and/or models were thought necessary and useful. Finally, capabilities with regard to model recalibration are limited, simply because of existing data limitations, availability of explanatory variables, and intersection representativeness across States. When these limitations are thought to be critical they are identified and discussed.

The technical strategy applied in this research effort is now described. Each of the strategies represents different possible end uses of the models, influenced by the stated guiding philosophical principles.

AADT Models: One set of models represents intersection crash models that forecast crashes in frequency-per-year based on minor and major road AADT-only. There are no other independent variables in these models. The intended use of these models is to provide a baseline crash forecast, which can then be modified with AMFs representing the effects of various geometric, roadside, and other relevant safety-related factors. The sample available for calibrating these models was much larger than the sample available for calibrating full models that, in a sense, partly compensates for the loss of statistical precision resulting from the omission of variables other than AADT.

Full Models: Another set of models represents statistical models with a full set of explanatory variables, including major and minor road AADT. These models are meant to provide a fuller understanding of the geometric, roadside, and operational features of intersections that influence on crashes. Another use might be to develop or infer additional crash modification factors for the various types of intersections examined in this research.

AMFs: A final set of "models" represents estimated effects of various geometric, roadside, and operational features. These provide a complement to the AADT models. The intended use of the AMFs is to provide percentage corrections to expected crash frequencies that result from the application of various crash countermeasures. AMFs represent a fairly intuitive approach to evaluating safety countermeasures, and are handled rather simply in the IHSDM.

When comparing and refining the three types models, several GOF measures were used in addition to inspection of model coefficients, collection of explanatory variables, and t-statistics and their associated p-values. Numerous measures are relied upon to avoid basing decisions on one single measure. Unfortunately, there is no one single criterion that dominates to the point of rendering the remaining measures as invalid or unimportant. It is through the assessment of many measures that a "best" model is chosen, and it is not always a clear winner.

3.1.1 Model Functional Forms

The negative binomial model form, which is identical to that used in previous efforts, was used to provide the best fit to the data.(1,2) The following model form and error distribution were assumed to represent the underlying phenomenon:

AADT Only Models

Equation 12. The mean number of accidents to be expected at site I in a given time period, Y tophat subscript I, equals exponent of the sum of the estimated intercept term, alpha, plus estimated coefficient beta subscript 1 times AADT subscript 1 plus estimated coefficient beta subscript 2 times AADT subscript 2. (12)

where

Y tophat = the mean number of accidents to be expected at site i in a given time period;

Alpha = the estimated intercept term; and

Beta1 Beta2, estimated coefficients.

Fully Parameterized Models

The following model form and error distribution were assumed to represent the underlying phenomenon:

Equation 13. Y tophat subscript I equals AADT subscript 1 raised to the beta subscript 1 power times AADT subscript 2 raised to the beta subscript 2 power, times the exponent of the sum of alpha plus the sum from J equals 3 to N of beta subscript IJ times the values of the non-traffic highway variables at site I during that time period, X subscript IJ. (13)

where

Y tophat = the mean number of accidents to be expected at site i in a given time period;

Alpha = the estimated intercept term;

Xi1, Xi2....Xin, = the values of the non-traffic highway variables at site i during that time period; and

Betai1 Betai2.....Betain, = estimated coefficients.

Equation 14. The estimated variance of the mean accident rate, Var open bracket M closed bracket, equals the estimated mean accident rate from the model, E open bracket M closed bracket, plus the estimated overdispersion parameter, K, times E open bracket M closed bracket squared. (14)

where

Var{m} = the estimated variance of the mean accident rate;

E{m} = the estimated mean accident rate from the model; and

K = the estimated overdispersion constant.

3.1.2 Goodness-of-Fit Evaluation

Four GOF measures were used in the model selection process (refer to chapter 2 for a description of the GOF measures.). A fifth approach to evaluating the GOF and in particular the suitability of alternate model forms was the Cumulative Residuals (CURE) method, proposed by Hauer and Hauer and Bamfo, in which the cumulative residuals (the difference between the actual and fitted values for each intersection) are plotted in increasing order for each covariate separately.(8,9) The graph shows how well the model fits the data with respect to each individual covariate. Figure 19 illustrates the CURE plot for the covariate AADT1 for the total accidents for the selected AADT-only model for Type III intersections (presented in table 142). The indication is that the fit is very good for this covariate in that the cumulative residuals oscillate around the value of zero and lie between the two standard deviation boundaries. Figure 20 is a CURE plot for an alternate model. Clearly, the alternate model cannot be judged to be an improvement over the selected model. Appendix D contains CURE plots for the TOTACC AADT models for all intersection types.

Figure 19. CURE Plot for Type III TOTACC AADT Model. Graph. This figure illustrates the CURE plot for the covariate AADT1 for the total accidents for the selected AADT-only model for Type III intersections. It plots adjusted cumulative residuals against two standard deviations. Major AADT is graphed on the X axis from 0 to 80,000 (in increments of 10,000), and cumulative residuals are graphed on the Y axis from negative 25 to 25. All adjusted cumulative residuals oscillate around the value of 0 and lie between the two standard deviation boundaries, which extend from negative 22 to positive 22 cumulative residuals.

Figure 19. CURE Plot for Type III TOTACC AADT Model

Figure 20. CURE Plot for Type III TOTACC AADT Model Using the CURE Method: Alternate Model. Graph. This figure illustrates the CURE plot for an alternate model. It plots adjusted cumulative residuals against two standard deviations. Major AADT is graphed on the X axis from 0 to 80,000 (in increments of 10,000), and cumulative residuals are graphed on the Y axis from negative 25 to 25. Most adjusted cumulative residuals oscillate around the value of 0 and lie between the two standard deviation boundaries, which extend from negative 22 to positive 22 cumulative residuals. There is one exception; one adjusted cumulate residual falls outside the standard deviation boundary at coordinates 62,000,12.

Figure 20. CURE Plot for Type III TOTACC AADT Model Using the CURE Method: Alternative Model

Now that the model's end uses, guiding research philosophies, and technical modeling strategy have been described, the details of the technical modeling efforts are presented and discussed. It is useful to first describe the data that were used in the model recalibration efforts, and to identify any difficulties, anomalies, and peculiar circumstances that needed to be remedied in the effort.

3.2 DESCRIPTION OF DATA

Different variables were used in developing statistical models for Types I and II compared to Types III, IV, and V. Although average daily traffic variables are common to all models, in general there were a larger number of variables available for estimation of model Types III, IV, and V. The abbreviation employed in the modeling efforts and their descriptions are provided in the following section.

3.2.1 Summary of Datasets

The data used for recalibration were obtained from three sources. The first two sets were identical to the data used for the validation exercise described in chapter 2. The first set was the original calibration data used by Vogt and Bared from Minnesota and Vogt from California and Michigan.(1,2) Additional years of accident and traffic data were obtained for those sites which did not experience a change in major variables, such as traffic control or number of legs. There were primarily minor differences in the summary statistics between those calculated on the available data and those stated in the reports, particularly for the vertical curvature variables for Type V sites. However, existing differences are sufficiently minor that further clarification was not necessary. The accident data obtained for the original sites included data for both the original and additional years. Differences were found in the accident counts between the original data obtained and this new dataset for the original years. Again, although small differences exist, their causes are unknown and these discrepancies were small enough that the data could confidently be used for recalibration. The second source of data was for those sites selected in Georgia to provide and an independent set of validation data. The third source of data was the California HSIS database. This data set was acquired to increase the size of the recalibration datasets with the aim of providing improved models with smaller standard errors of parameter estimates. These data were collected with a minimum amount of effort with assistance from the HSIS staff. However, as site visits were not conducted, fewer variables were available for these sites. Table 108 summarizes the sources of data used for recalibrating Models I to V.

Table 108. Sources of Data

State
Years of Data Available
No. of Sites
No. of Total (Injury Accidents)
Type I
Type II
Type III
Type IV
Type V
Type I
Type II
Type III
Type IV
Type V
Minnesota
1985-98
270
250
N/A4
N/A4
N/A4
2029
(788)
1892
(878)
N/A4
N/A4
N/A4
California1
1991-98
1432
748
294
222
75
6494
(2978)
6063
(3058)
2136
(847)
1956
(899)
1159
(370)
California2
1993-98
N/A4
N/A4
60
54
18
N/A4
N/A4
427
(196)
478
(268)
507
(200)
Michigan3
1993-97
N/A4
N/A4
24
18
31
N/A4
N/A4
248
(63)
277
(92)
1262
(159)
Georgia
1996-97
116
108
52
52
51
295
(110)
255
(142)
124
(56)
222
(104)
489
(118)
Total  
1818
1106
430
346
124
8818
(3908)
8210
(4078)
2935
(1162)
2933
(1363)
3417
(847)

1 These data come from the California HSIS database and do not include variables, such as vertical curvature, not available electronically in that database

2 Only the original sites were used to develop the base models for Types III, IV, and V, and only the California HSIS sites were used to develop the full models

3 For Type V, Only 1996-97 injury accidents were available

4 N/A: not available

In this section, summary statistics are provided for the data available for recalibrating the full models (i.e. models with explanatory variables other than traffic volumes). For model Types III, IV, and V, the California HSIS sites were not included due to the limited availability of variables relevant to these models.

It is also appropriate and useful to examine which variables strongly correlate positively or negatively with crashes and which potential independent variables are correlated to one another. These statistics are also provided in this section of the report.

3.2.2 Type I

A summary of the full data for Type I intersections is shown in table 109. This dataset includes the original sites in Minnesota, with the additional years of accident and traffic data, the Georgia sites and the California HSIS sites. Some of the Minnesota sites experienced changes in some design feature or location information during the 1990-98 period and were not included in the analysis. Note that some variables are not available in the data for the Minnesota sites and California sites. The frequency column indicates the number of sites for which the information was available. Summary statistics by State are available in appendix C.

Table 109. Summary Statistics for Type I Sites

Variables
Frequency
Mean
Median
Minimum
Maximum
TOTACC per year
1818
0.6074
0.3750
0
6.75
INJACC per year
1818
0.2660
0.1250
0
4.13
AADT1
1818
6011
4475
401
35750
AADT2
1818
492
270
100
10001
RT MAJ Total
1818
N/A1
0
1563 (86%)
1
255 (14%)
RT MIN Total
1818
N/A1
0
1770 (97.4%)
1
48 (2.6%)
LT MAJ Total
1818
N/A1
0
1382 (76%)
1
436 (24%)
LT MIN Total
1818
N/A1
0
1804 (99.2%)
1
14 (0.8%)
MEDIAN Total
1818
N/A1
0
1738 (95.6%)
1
80 (4.4%)
TERRAIN Total
1548
N/A1
Flat
568 (31.2%)
Rolling
547 (30.1%)
Mountainous
433 (23.8%)
SPD1
381
50.89
55
23
55
DRWY1
386
1.38
1
0
8
HAZRAT1
386
2.56
2
1
7
HAU
386
-1.451
0
-90
85.1
SHOULDER1
1547
4.75
4
0
16
VCI1
386
0.477
0
0
14.0
HI1
386
1.6553
0
0
29.0

1 N/A: not available

Table 110 shows correlation statistics and p-values that indicate the association between crash counts and the independent variables for type I intersections. Table 111 shows correlations between the independent variables. Only those correlations that are significant at the 90 percent level are shown.

As expected, major and minor road AADTs correlate positively with crashes. Turning lanes on the major and minor roads are also positively correlated with crashes, although this correlation is much less than that of vehicle volumes and the correlation for right-turn lane on major roads is not significant. Surprisingly, terrain and posted speed are negatively correlated with crashes, meaning that areas with rolling or mountainous terrain experience a lower crash risk than flatter terrains and that higher speeds are associated with fewer crashes. This counterintuitive result may arise because, as shown in Appendix C, Georgia sites have higher accident frequencies than California and Minnesota sites, as well as lower average posted speeds and a higher percentage of sites in rolling or mountainous terrain. With the presence of a median, VCI1 and HI1 were positively correlated with crashes, while HAU was negatively correlated with crashes although this correlation was not as strong. Shoulder width and number of driveways were not significantly correlated with crashes.

Table 110. Correlation Between Crashes and Independent Variables for Type I Sites

Variables
TOTACC per YEAR
INJACC per YEAR
Corr.
p-value
Corr.
p-value
AADT1
0.426
0.000
0.402
0.000
AADT2
0.428
0.000
0.327
0.000
RT MAJ
0.030
0.202
0.005
0.841
RT MIN
0.116
0.000
0.106
0.000
LT MAJ
0.165
0.000
0.149
0.000
LT MIN
0.059
0.012
0.056
0.016
TERRAIN
-0.085
0.001
-0.101
0.000
MEDIAN
0.076
0.001
0.074
0.002
SPD11
-0.127
0.013
-0.065
0.205
DRWY11
0.030
0.558
0.020
0.694
HAU1
-0.072
0.157
-0.052
0.312
SHOULDER12
-0.020
0.427
0.013
0.619
VCI11
0.081
0.110
0.033
0.516
HI11
0.087
0.088
0.089
0.080

1 These variables are unknown for the California sites

2 These variables are unknown for the Minnesota sites

Table 111. Summary of Correlations for Independent Variables for Type I Sites

Variable
Positive Correlates1
Negative Correlates1
AADT1
AADT2, RT MIN, LT MAJ, MEDIAN, SHOULDER1
VCI1, HI1, TERRAIN
AADT2
AADT1, RT MAJ, RT MIN, LT MAJ, LT MIN, MEDIAN, HI1
TERRAIN
RT MAJ
AADT2, RT MIN, LT MAJ, LT MIN, SPD1, SHOULDER1
HAZRAT1, VCI1, HI1
RT MIN
AADT1, AADT2, RT MAJ, LT MAJ, LT MIN, MEDIAN, TERRAIN
 
LT MAJ
AADT1, AADT2, RT MAJ, RT MIN, LT MIN, MEDIAN, SHOULDER1
TERRAIN, HAZRAT1, VCI1
LT MIN
AADT2, RT MAJ, RT MIN, LT MAJ, MEDIAN
 
MEDIAN
AADT1, AADT2, RT MIN, LT MAJ, LT MIN, VCI1
TERRAIN, SPD1, SHOULDER1
TERRAIN
RT MIN, HAZRAT1, HI1
AADT1, AADT2, LT MAJ, MEDIAN, SPD1, SHOULDER1
SPD1
RT MAJ, SHOULDER1
MEDIAN, TERRAIN, NODRWAY, HAZRAT1, VCI1, HI1
DRWY1
HI1
SPD1
HAZRAT1
TERRAIN, VCI1, HI1
RT MAJ, LT MAJ, SPD1
HAU
   
SHOULDER1
AADT1, RT MAJ, LT MAJ, SPD1
MEDIAN, TERRAIN, VCI1
VCI1
MEDIAN, HAZRAT1
AADT1, RT MAJ, LT MAJ, SPD1, SHOULDER1
HI1
AADT2, TERRAIN, DRWY1, HAZRAT1
 

1 Only those correlations are shown for which p-values are less than 0.10.

2 Not all variables are available for Minnesota or California sites

3.2.3 Type II

A summary of the full data for Type II intersections is shown in table 112. This dataset includes the original sites in Minnesota, with additional years of accident and traffic data, the Georgia sites and the California HSIS sites. Some of the Minnesota sites experienced changes in some design feature or location information during 1990-98 and were not included in the analysis. Note that some variables are not available in the data for the Minnesota sites and California sites. The frequency column indicates the number of sites for which the information was available. Summary statistics by State are available in appendix C.

Table 112. Summary Statistics for Type II Sites

Variables
Frequency
Mean
Median
Minimum
Maximum
TOTACC per year
1106
0.9227
0.5357
0
7.13
INJACC per year
1106
0.4665
0.2500
0
4.75
AADT1
1106
5487
4245
407
38126
AADT2
1106
532
344
100
7460
RT MAJ Total
0
1
1106
911 (82.4%)
195 (17.6%)
N/A1
RT MIN Total
0
1
1106
1080 (97.6%)
26 (2.4%)
N/A1
LT MAJ Total
0
1
1106
883 (79.8%)
223 (20.2%)
N/A1
LT MIN Total
0
1
1106
1105 (99.9%)
1 (0.1%)
N/A1
MEDIAN Total
0
1
1106
1069 (96.7%)
37 (3.3%)
N/A1
TERRAIN Total
Flat
Rolling
Mountainous
856
520 (47%)
238 (21.5%)
98 (8.9%)
N/A1
SPD1
355
52
55
30
55
DRWY1
358
0.83
0
0
6
HAZRAT1
358
2.45
2.00
1
6
HAU
358
0.364
0
-120
150
SHOULDER1
855
5.426
6
0
16
VCI1
358
0.43
0.05
0
8
HI1
358
0.896
0
0
14.553

1 N/A: not available

Table 113 shows correlation statistics and p-values that indicate the association between crash counts and the independent variables for Type II intersections. Table 114 shows correlations between the independent variables. Only those correlations that are significant at the 90 pecent level are shown. Note that some variables are not included in the data for the Minnesota and California sites.

As expected, major and minor road AADTs correlate positively with crashes. Right-turn lanes on the major roads were negatively correlated with crashes, while right-turn lanes on the minor roads were positively correlated with crashes. Left-turn lanes on the major roads were positively correlated with crashes, however left-turn lanes on the minor roads were not significantly correlated with crashes. Again, terrain and posted speed are negatively correlated with crashes, meaning that areas with rolling or mountainous terrain experience a higher crash risk than flatter geographies and that higher speeds are associated with less crashes. Presence of a median, number of driveways, HI1, and roadside hazard rating on the major roads were all positively correlated with crashes. Intersection angle (HAU), shoulder width, and VCI1 were not significantly correlated with crashes.

Table 113. Correlation Between Crashes and Independent Variables for Type II Sites

Variables
TOTACC per YEAR
INJACC per YEAR
Corr.
p-value
Corr.
p-value
AADT1
0.443
0.000
0.384
0.000
AADT2
0.434
0.000
0.425
0.000
RT MAJ
-0.133
0.000
-0.126
0.000
RT MIN
0.111
0.000
0.105
0.000
LT MAJ
0.258
0.000
0.265
0.000
LT MIN
0.027
0.364
0.028
0.353
TERRAIN
-0.103
0.003
-0.115
0.001
MEDIAN
0.088
0.003
0.060
0.046
SPD11
-0.246
0.000
-0.184
0.001
DRWY11
0.251
0.000
0.197
0.000
HAZRAT11
0.152
0.004
0.101
0.057
HAU1
-0.041
0.444
0.007
0.895
SHOULDER13
0.008
0.821
-0.001
0.970
VCI11
0.029
0.580
0.046
0.390
HI11
0.086
0.106
0.123
0.020

1 These variables are unknown for the California sites

2 These variables are unknown for the Minnesota sites

Table 114. Summary of Correlations for Independent Variables for Type II Sites

Variable1
Positive Correlates2
Negative Correlates2
AADT1
AADT2, RT MIN, LT MAJ, LT MIN, MEDIAN, DRWY1, HAZRAT1, SHOULDER1
RT MAJ
AADT2
AADT1, RT MIN, LT MAJ, MEDIAN, TERRAIN, DRWY1, HAZRAT1
SPD1
RT MAJ
RT MIN, TERRAIN, SPD1, SHOULDER1
AADT1, DRWY1, HAZRAT1, VCI1, HI1
RT MIN
AADT1, AADT2, RT MAJ, LT MAJ, LT MIN
 
LT MAJ
AADT1, AADT2, RT MIN, LT MIN, MEDIAN, TERRAIN, HAZRAT1, SHOULDER1, VCI1, HI1
 
LT MIN
AADT1, RT MIN, LT MAJ, MEDIAN
 
MEDIAN
AADT1, AADT2, LT MAJ, LT MIN, TERRAIN
SHOULDER1
TERRAIN
AADT2, RT MAJ, LT MAJ, MEDIAN, HAZRAT1, VCI1, HI1
SPD1, SHOULDER1
SPD1
RT MAJ, SHOULDER1
AADT2, TERRAIN, DRWY1, HAZRAT1, VCI1, HI1
DRWY1
AADT1, AADT2, HAZRAT1, VCI1, HI1
RT MAJ, SPD1
HAZRAT1
AADT1, AADT2, LT MAJ, TERRAIN, DRWY1, VCI1, HI1
RT MAJ, SPD1, HAU
HAU
 
HAZRAT1
SHOULDER1
AADT1, RT MAJ, LT MAJ, SPD1
MEDIAN, TERRAIN, HAZRAT1
VCI1
LT MAJ, TERRAIN, DRWY1, HAZRAT1, HI1
RT MAJ, SPD1
HI1
LT MAJ, TERRAIN, DRWY1, HAZRAT1, VCI
RT MAJ, SPD1

1 Not all variables are available for Minnesota or California sites

2 Only those correlations are shown for which p-values are less than 0.10

3.2.4 Type III

A summary of the full data for Type III intersections is shown in table 115. In total, 42 variables were available for model development. The HSIS California data were excluded in developing Type III full models because this data set has only a few variables (turning lanes, median, terrain, etc) of relevance. This left the California and Michigan sites from the original study, with the additional years of accident data, for inclusion in the database. Some California sites experienced changes in some design features during 1996-98. For these, only 1993-95 data were used. As before the frequency column indicates the number of sites for which the information was available.

Table 115. Summary Statistics for Type III Sites

Variables
Frequency
Mean
Median
Minimum
Maximum
TOTACC per year
136
1.35
0.80
0.00
10.60
INJACC per year
136
0.55
0.33
0.00
4.00
AADT1
136
13011
12100
2360
33333
AADT2
136
709
430
15
9490
MEDTYPE1 Total
136
N/A1
No Median
69(50.7%)
Painted
45(33.1%)
Curbed
14(10.3%)
Other
8(5.9%)
MEDWIDTH1
136
12.6
6
0
63
HAU
136
1.3
0
-65
90
HAZRAT1 Total
136
N/A1
1
16(11.8%)
2
58(42.6%)
3
26(19.1%)
4
25(18.4%)
5
8(5.9%)
6
2(1.5%)
7
1(0.7%)
HAZRAT2 Total
52
N/A1
1
0(0%)
2
2(4.0%)
3
20(40.0%)
4
16(32.0%)
5
6(12.0%)
6
6(12.0%)
7
2(4.0%)
COMDRWY1
136
1.5
0
0
14
RESDRWY1
136
1.0
0
0
7
DRWY1
136
2.5
1.0
0.0
15.0
NoCOMDRWY2
52
0.4
0
0
3
RESDRWY2
52
0.6
0
0
6
DRWY2
52
1.0
1.0
0.0
6.0
SPD1
136
52.5
55
30
65
SPD2
136
33.7
35
15
55

1 N/A: not available

Table 115. Summary Statistics for Type III Sites (Continued)

Variables
Frequency
Mean
Median
Minimum
Maximum
LIGHT Total
136
N/A1
0
97(71.3%)
1
39(28.7%)
TERRAIN1 Total
136
N/A1
Flat
83(61.0%)
Rolling
42(30.9%)
Mountainous
11(8.1%)
TERRAIN2 Total
52
N/A1
Flat
24(17.6%)
Rolling
21(15.4%)
Mountainous
7(5.1%)
RTLN1 Total
136
N/A1
0
108(79.4%)
1
28(20.6%)
LTLN1 Total
136
N/A1
0
48(35.3%)
1
88(64.7%)
RTLN2 Total
136
N/A1
0
117(86.0%)
1
19(14.0%)
LTLN2 Total
136
N/A1
0
131(96.3%)
1
5(3.7%)
HI1
136
1.26
0.00
0
14.29
HEI1
136
2.01
0.73
0
26.63
GRADE1
136
1.0
0.7
0.0
5.9
GRADE2
52
1.5
1.2
0.0
4.7
VEI1
136
0.9
0.6
0.0
6.7
VI2
52
4.0
2.8
0.0
24.0
LEGACC1
52
0.0
0.0
0.0
1.0
LEGACC2
52
0.1
0.0
0.0
1.0
SHOULDER1
52
4.0
4.0
0.0
10.0
PKTRUCK
84
9.15
7.79
1.18
28.16
PKTURN
84
6.68
4.28
0.27
53.09
PKLEFT
84
3.28
2.16
0.13
25.97

1 N/A: not available

Table 115. Summary Statistics for Type III Sites (Continued)

Variables
Frequency
Mean
Median
Minimum
Maximum
PKLEFT1
84
1.47
0.69
0.00
21.29
PKLEFT2
84
55.31
60.29
0.00
100.00
SD1
136
1515
2000
500
2000
SDL2
136
1418
1510
40
2000
SDR2
136
1428
1555
80
2000

Table 116 shows correlation statistics and p-values that indicate the association between crash counts and the independent variables for Type III intersections. Table 114 shows correlations between the independent variables. Only those correlations that are significant at the 90 percent level are shown.

Major and minor road AADTs correlate positively with crashes as expected. Peak turning movement volumes also correlate with crashes, both positively and negatively. PKTURN, PKLEFT, and PKLEFT1 correlate positively with crashes, while PKTRUCK and PKLEFT2 correlate negatively with crashes. According to table 114, PKTRUCK correlates negatively with the AADT variables. This suggests that the negative correlation of crashes with PKTRUCK may, in part, be a consequence of the positive correlation of crashes with AADT variables. PKLEFT1 and PKLEFT2 are also negatively correlated with each other. There are several variables for which the correlation results are unexpected. Roadside hazard rating on major and minor roads, number of residential driveways on major and minor roads, posted speed limits on major and minor roads, terrain on major roads, shoulder width on major roads, "LIGHT," and the presence of left-and right-turn lane on minor roads, as well as other variables are correlated with crashes in the opposite direction to that expected, although many of these correlations are insignificant.

Table 116. Correlation Between Crashes and Independent Variables for Type III Sites

Variables
TOTACC per YEAR
INJACC per YEAR
Corr.
p-value
Corr.
p-value
AADT1
0.3330
0.0001
0.2943
0.0005
AADT2
0.4829
0.0000
0.3606
0.0000
MEDWDTH1
-0.0774
0.3703
-0.0051
0.9534
HAU
0.1190
0.1677
0.1917
0.0254
COMDRWY1
0.3959
0.0000
0.1765
0.0398
RESDRWY1
-0.0697
0.4201
-0.1211
0.1603
DRWY1
0.2842
0.0008
0.0854
0.3229
COMDRWY2
0.0044
0.9756
0.0486
0.7321
RESDRWY2
-0.2342
0.0947
-0.2062
0.1425
DRWY2
-0.1956
0.1647
-0.1416
0.3168
SPD1
-0.3299
0.0001
-0.1184
0.1696
SPD2
-0.0675
0.4352
0.0519
0.5483
LIGHT
0.2882
0.0007
0.1307
0.1295

Table 116 . Correlation Between Crashes and Independent Variables for Type III Sites (Continued)

Variables
TOTACC per YEAR
INJACC per YEAR
Corr.
p-value
Corr.
p-value
L1RT
0.0118
0.8915
0.0344
0.6911
L1LT
-0.1511
0.0791
0.0192
0.8243
L3RT
0.2298
0.0071
0.1717
0.0456
L3LT
0.2025
0.0181
0.2373
0.0054
HI1
0.0309
0.7208
0.0615
0.4771
HEI1
0.0052
0.9520
0.1628
0.0583
GRADE1
0.0027
0.9748
0.0485
0.5751
GRADE2
0.0968
0.4949
0.1977
0.1601
VEI1
0.1534
0.0746
0.1247
0.1481
VI2
-0.1039
0.4633
-0.0831
0.5582
LEGACC1
-0.0721
0.6116
-0.1020
0.4719
LEGACC2
0.2099
0.1353
-0.0129
0.9278
SHOULDER1
0.1392
0.3249
-0.0140
0.9216
PKTRUCK
-0.1943
0.0766
-0.1205
0.2749
PKTURN
0.2617
0.0162
0.2527
0.0204
PKLEFT
0.2304
0.0350
0.2296
0.0357
PKLEFT1
0.2744
0.0115
0.2479
0.0230
PKLEFT2
-0.1610
0.1436
-0.0994
0.3685
SD1
-0.0752
0.3843
-0.0003
0.9970
SDL2
-0.0633
0.4642
-0.0300
0.7284
SDR2
-0.0585
0.4986
-0.0214
0.8043

Table 117. Summary of Correlations for Independent Variables for Type III Sites

Variable
Positive Correlates1
Negative Correlates1
AADT1
L1RT, L1LT
MEDTYPE2, PKTRUCK,PKLEFT2, SDL2
AADT2
L1RT, L3RT, L3LT, PKTURN, PKLEFT,PKLEFT1, SHOULDER1,
SPD1, PKTRUCK
MEDWDTH1
HAU, SPD1, SPD2, L1RT, L1LT, PKTRUCK, SHOULDER1, SDR2
COMDRWY1, RESDRWY1, DRWY1, LIGHT, TERRAIN, HI1, GRADE1, VI2,
HAU
MEDWDTH1, PKTRUCK, LEFACC2,
MEDTYPE2, RESDRWY1, DRWY2,
HAZRAT1
HAZRAT2, SPD1, SPD2, TERRAIN1, L1LT, GRADE1, VEI1
COMDRWY1, RESDRWY1, DRWY1, LIGHT, L1LT, SDR2
DRWY1
COMDRWY1, RESDRWY1, COMDRWY2, DRWY2, LIGHT, PKTURN, HI1
MEDTYPE1, MEDTYPE2, MEDTYPE3, HAZRAT1, SPD1, SPD2, L1RT, L1LT, PKTRUCK, PKLEFT2, SDL3, SDR3
SPD1
MEDTYPE1,MEDTYPE3, MEDWDTH1, SPD2, TERRAIN1, L1RT, L1LT, PKTRUCK, LEGACC2, SD1, SDL2, SDR2
AADT2, COMDRWY1, RESDRWY1, DRWY1, COMDRWY2, DRWY2, LIGHT, HI1, GRADE2, VEI1
SPD2
MEDTYPE1, MEDWDTH1, HAZRAT1, SPD1, TERRAIN1, L1RT, L1LT, L3LT,
COMDRWY1, DRWY1, LIGHT
LIGHT (no=0, yes=1)
COMDRWY1, PKTURN, HI1, LEFACC1, DRWY1, PKLEFT, PKLEFT1
MEDTYPE2, MEDTYPE3, MEDWDTH1, HAZRAT1, SPD1, SPD2, L1LT, PKTRUCK, SD1, SDR2
TERRAIN1
MEDTYPE1, HAZRAT1, HAZRAT2, SPD1, SPD2, L1RT, GRADE1, GRADE2, VEI1, VI2
SD1, SDL2, SDR2
L1RT
AADT1, AADT2, MEDWDTH1, SPD1, SPD2, TERRAIN1, L1LT, L3RT, L3LT, GRADE1, LEFACC2, SHOULDER1
HAZRAT2, COMDRWY1, RESDRWY1, DRWY1, COMDRWY2, GRDE2, TERRAIN2
L1LT
AADT1, MEDTYPE1, MEDTYPE2, MEDWDTH1, HAZRAT1, SPD1, SPD2, L1RT, L3LT, SD1, SDR3
HAZRAT2, COMDRWY1, RESDRWY1, DRWY1, LIGHT, TERRAIN2
L3RT
AADT2, L1RT, L3LT, PKTURN, SHOULDER1, PKTURN, PKLEFT, PKLEFT1
HAZRAT2, TERRAIN2
L3LT
AADT2, MEDTYPE1, SPD2, L1RT, L1LT, L3RT, PKTURN, PKLEFT, PKLEFT1
HAZRAT2
PKTRUCK
MEDTYPE1, MEDTYPE3, MEDWDTH1, HAU, SPD1, SPD2, SD1, SDL2, SDR2
AADT1, AADT2, COMDRWY1, RESDRWY1, DRWY1, LIGHT, HI1, VEI1,
PKTURN
AADT2, LIGHT, L3RT, L3LT, PKLEFT, PKLEFT1
 
VEI1
AADT1, HAZRAT1, TERRAIN1, HI1, GRADE1,
SPD1, PKTRUCK, SD1, SDL2, SDR2
HEI1
MEDTYPE1, HI, VI2
 
GRADE1
MEDTYPE1, HAZRAT1, TERRAIN1, L1RT, HI1, VEI1
MEDWDTH1, SD1, SDL2, SDR2
SDL2
SPD1, PKTRUCK, SD1, SDR2
AADT1, RESDRWY1, DRWY1, TERRAIN1, TERRAIN2, HI1, GRADE1, GRADE2, VEI1, LEGACC2
SDR2
MEDWDTH1, SPD1, L1LT, PKTRUCK, SD1, SDL3
HAZRAT1, HAZRAT2, LIGHT, TERRAIN1, HI1, GRADE1, GRADE2, VEI1, DRWY1

1Only those correlations are shown for which p-values are less than 0.10

3.2.5 Type IV

A summary of the full data for type IV intersections is shown in table 118. In total, 53 variables were available for model development. The HSIS California data were again excluded because of a lack of sufficient variables (turning lanes, median, terrain, etc.) of relevance. Instead, the California and Michigan sites from the original study, with the additional years of accident data were included in the database. Some California sites experienced changes in some design features during 1996-98. For these, only 1993-95 data were used. As before, frequency indicates the number of sites for which the information was available.

Table 118. Summary Statistics for Type IV Sites

Variables
Frequency
Mean
Median
Minimum
Maximum
TOTACC per YEAR
124
2.0
1.4
0.0
10.8
INJACC per YEAR
124
0.9
0.5
0.0
5.7
AADT1
124
12881
11496
3150
73799
AADT2
124
621
430
21
2990
MEDTYPE on major Total
124
N/A1
0: No Median
70(56.5%)
1: Painted
27(21.8%)
2: Curbed
22(17.7%)
3: Other
5(4.0%)
MEDTYPE on minor Total
52
N/A1
0: No Median
52(100%)
MEDWDTH1
124
16.1
6.5
0
60
MEDWDTH2
52
0.0
0
0
1
SHOULDER1
52
4.2
4
2
6
SHOULDER2
52
0.3
0
0
2
L1RT Total
124
N/A1
0
69(55.6%)
1
20(16.1%)
2
35(28.2%)
L3RT Total
124
N/A1
0
72(58.1%)
0
13(10.5%)
2
39(31.5%)
L3LT Total
124
N/A1
0
122(98.4%)
1
2(1.6%)
LEGACC1 Total
N/A1
0
52
0
49(94.2%)
1
3(5.8%)
LEGACC2 Total
52
N/A1
0
49(94.2%)
1
3(5.8%)
HAZRAT1
124
N/A1
1
24(19.4%)
2
43(34.7%)
3
32(25.8%)
4
21(16.9%)
5
2(1.6%)
6
2(1.6%)
7
0(0%)

1 N/A: not available

Table 118 . Summary Statistics for Type IV Sites (Continued)

Variables
Frequency
Mean
Median
Minimum
Maximum
HAZRAT2
52
N/A1
1
0(0%)
2
7(13.5%)
3
15(28.8%)
4
16(30.8%)
5
12(23.1%)
6
2(3.8%)
7
0(0%)
COMDRWY1
124
0.6
0
0
12
RESDRWY1
124
0.7
0
0
7
DRWY1
124
1.3
0
0
15
COMDRWY2
52
0.4
0
0
4
RESDRWY2
52
0.4
0
0
3
DRWY2
52
0.8
0
0
6
LIGHT Total
124
N/A1
0
87(70.2%)
1
37(29.8%)
TERRAN1Total
124
N/A1
Flat
90(72.6%)
Rolling
25(20.2%)
Mountainous
9(7.3%)
TERRAN1Total
52
N/A1
Flat
19(36.5%)
Rolling
27(51.9%)
Mountainous
6(11.5%)
VEI1
124
0.87
0.35
0.00
12.50
VCEI1
124
0.63
0.00
0.00
12.50
VI1
124
0.62
0.00
0.00
12.50
VCI1
124
0.43
0.00
0.00
12.50
VEI2
52
3.05
2.84
0.32
10.18
VCEI2
52
2.97
2.31
0.00
11.36
VI2
52
2.62
2.08
0.00
9.66
VC12
52
2.08
1.02
0.00
12.50
GRADE1
124
0.94
0.71
0.00
5.80
GRADE2
51
1.65
1.48
0.60
3.71
HI
124
0.92
0.00
0.00
7.33
HEI
124
3.28
0.60
0.00
233.33
HAU
124
1.5
0
-50
55
SPD1
124
55.6
55
25
65

1 N/A: not available

Table 118 . Summary Statistics for Type IV Sites (Continued)

Variables
Frequency
Mean
Median
Minimum
Maximum
SPD2
124
34.7
35
25
55
PKTRUCK
72
10.95
8.36
1.75
37.25
PKTHRU1
72
94.41
96.95
67.77
100.00
PKTURN
72
9.47
6.56
0.00
48.52
PKLEFT
72
4.80
3.08
0.00
25.26
PKLEFT1
72
2.78
1.51
0.00
13.96
PKTHRU2
72
15.69
10.82
0.00
68.09
PKLEFT2
72
38.89
36.66
0.00
100.00
SD1
124
1399
1332
400
2000
SDL2
124
1314
1262
324
2000
SDR2
124
1329
1354
215
2000

1 N/A: not available

Table 119 shows correlation statistics and p-values that indicate the association between crash counts and the independent variables for Type IV intersections. Table 120 shows correlations between the independent variables. Only those correlations that are significant at the 90 percent level are shown.

Major and minor road AADTs correlate positively with crashes, as expected. Peak turning movements also correlate with crashes, both positively and negatively. There are several variables for which the correlation results are contrary to expectations. Shoulder width on the road, right-and left-turn lane on minor roads, acceleration lane on major roads, residential driveway and total driveway on minor roads, light, terrain on major and minor roads, vertical curves on major and minor roads, horizontal curves on major roads, absolute grades on major and minor roads, intersection angle, posted speed limit on major roads, and others are correlated with crashes in the opposite direction than expected, although many of these correlations are insignificant. For example, median width on major road is insignificant with a counterintuitive sign. However, as table 120 shows, there is a negative correlation between median width on major roads and median types, the result of which is that median type is skewing the effect of median width at Type IV intersections.

Table 119. Correlation Between Crashes and Independent Variables for Type IV Sites

Variables
TOTACC per YEAR
INJACC per YEAR
Corr.
p-value
Corr.
p-value
AADT1
0.2258
0.0117
0.2285
0.0107
AADT2
0.2600
0.0035
0.1594
0.0770
MEDWDTH1
0.0314
0.7289
0.0572
0.5277
MEDWDTH2
-0.0104
0.9418
-0.0657
0.6434
SHOULDER1
-0.1631
0.2481
-0.1040
0.4633
SHOULDER2
0.2089
0.1372
0.2209
0.1155
L1RT
-0.0084
0.9267
0.0608
0.5026
L1LT
-0.0695
0.4432
0.0738
0.4152
L3RT
0.0350
0.6999
0.0995
0.2714
L3LT
0.1428
0.1137
0.1929
0.0319
LEGACC1
0.1633
0.2474
0.2323
0.0975
LEGACC2
-0.1092
0.4411
0.0000
1.0000
COMDRWY1
0.1017
0.2613
0.0942
0.2979
RESDRWY1
0.1547
0.0863
0.0015
0.9867
DRWY1
0.1569
0.0818
0.0596
0.5109
COMDRWY2
0.1900
0.1772
0.1732
0.2195
RESDRWY2
-0.2809
0.0437
-0.2474
0.0770
DRWY2
-0.0367
0.7963
-0.0283
0.8423
LIGHT
0.0592
0.5137
-0.0176
0.8459
VEI1
0.0099
0.9133
0.0373
0.6806
VCEI1
0.0765
0.3984
0.0698
0.4408
VI1
-0.0174
0.8476
0.0191
0.8332
VCI1
0.0151
0.8676
0.0490
0.5887
VEI2
-0.2156
0.1248
-0.0692
0.6257
VCEI2
-0.2626
0.0600
-0.0361
0.7994
VI2
-0.2665
0.0562
-0.0672
0.6360
VCI2
-0.2147
0.1263
-0.0506
0.7215
GRADE1
-0.0033
0.9709
0.0211
0.8161
GRADE2
-0.1825
0.1999
-0.0318
0.8245
HI1
-0.0329
0.7171
-0.0846
0.3503
HEI1
-0.0055
0.9519
-0.0581
0.5212
HAU
-0.1184
0.1905
-0.0892
0.3243
SPD1
-0.1839
0.0409
-0.0607
0.5033
SPD2
0.0301
0.7397
0.1964
0.0288
PKTRUCK
-0.3268
0.0051
-0.3369
0.0038
PKTHRU1
-0.3058
0.0090
-0.2324
0.0494
PKTURN
0.3242
0.0055
0.2544
0.0311
PKLEFT
0.3099
0.0081
0.2526
0.0323
PKLEFT1
0.3550
0.0022
0.3028
0.0097
PKTHRU2
0.1876
0.1145
0.1500
0.2086
PKLEFT2
-0.0492
0.6815
-0.0627
0.6006
SD1
-0.1331
0.1407
-0.1220
0.1770
SDL2
-0.1408
0.1187
-0.0849
0.3486
SDR2
-0.2826
0.0015
-0.1705
0.0583

Table 120. Summary of Correlations for Independent Variables for Type IV Sites

Variable
Positive Correlates1
Negative Correlates1
AADT1
MEDTYPE1, L1LT, SPD1, PKTHRU1, PKLEFT2
VCEI2, PKTRUCK, PKTURN, PKLEFT, PKLEFT1, PKTHRU2
AADT2
MEDWDTH1, MEDWDTH2, TERRAIN2, HEI1, HAU, PKTURN, PKLEFT, PKLEFT1, PKTHRU2
MEDTYPE1, GRADE1, PKTURCK, PKTHRU1, PKLEFT2
MEDWDTH1
AADT2, L1RT, L1LT, L3RT, HAZRAT1, HAU, SPD1, SPD2, PKTHRU1
MEDTYPE1, MEDTYPE2, HAZRAT2, COMDRWY1, RESDRWY1, COMDRWY2, RESDRWY2, DRWY1, DRWY2, LIGHT, TERRAIN1, VEI2, VCEI2, VI2, VCI2, PKTURN, PKLEFT, PKLEFT1
HAU
AADT2, MEDWDTH1, TERRAIN2
LIGHT
HAZRAT1
MEDTYPE1, MEDWDTH1, TERRAIN1, GRADE1, HI1,
MEDTYPE2, L1RT, L3RT, SD1, SDL2, SDR2, PKTRUCK, PKTHRU2
DRWY1
HAZRAT1, COMDRWY1, RESDRWY1, COMDRWY2, RESDRWY2, DRWY2, LIGHT, VI2, HEI1, PKTURN, PKLEFT, PKLEFT1
MEDTYPE2, MEDWDTH1, L1RT, L1LT, L3RT, SPD1, SPD2, PKTRUCK, PKTHRU1, SD1, SDL2, SDR2
SPD1
AADT1, MEDTYPE2, MEDWDTH1, SHOULDER2, L1RT, L1LT, L3RT, TERRAIN2, SPD2, PKTRUCK, PKTHRU1, SD1, SDL2, SDR2
COMDRWY1, RESDRWY1, DRWY1, COMDRWY2, DRWY2, LIGHT, HEI1, PKTURN, PKLEFT, PKLEFT1
SPD2
MEDWDTH1, L1RT, L1LT, L3RT, SPD1
HAZRAT2, COMDRWY1, RESDRWY1, DRWY1, RESDRWY2, DRWY2, LIGHT, VEI2, VCEI2, VI2, VCI2, HEI1
LIGHT (no=0,yes=1)
COMDRWY1, RESDRWY1, DRWY1, COMDRWY2, DRWY2, HEI1, PKTURN, PKLEFT, PKLEFT1
MEDTYPE2, MEDWDTH1, L1RT, L1LT, L2RT, HAU, SPD1, SPD2, PKTRUCK, PKTHRU1
TERRAIN1
MEDTYPE1, MEDWDTH2, SHOULDER2, L1LT, LEGACC1, HAZRAT1, GRADE1, HI1,
MEDWDTH1, PKTRUCK, PKTHRU2, PKLEFT2, SD1, SDL2, SDR2
L1RT
MEDTYPE2, MEDTYPE3, MEDWDTH1, L1LR, L3RT, LEGACC1, SPD1, SPD2, PKTRUCK, PKLEFT
HAZRAT1, RESDRWY1, DRWY1, LIGHT, TERRAIN2, GRADE1, GRADE2, HI1, PKTURN, PKLEFT, PKLEFT1
L1LT
AADT1, MEDTYPE1, MEDTYPE2, MEDWDTH1, L1RT, L3RT, TERRAIN1, SPD1, SPD2, PKTRUCK
COMDRWY1, RESDRWY1, DRWY1, RESDRWY2, DRWY2, LIGHT, VEI2, VCEI2, VI2, VCI2, GRADE2, HEI, PKTURN, PKLEFT, PKLEFT1
L3RT
MEDTYPE2, MEDWDTH1, SHOULDER2, L1LT, L1RT, SPD1, SPD2, PKTRUCK, PKTHRU1
MEDTYPE1, MEDTYPE3, HAZRAT1, HAZRAT2, COMDRWY1, RESDRWY2, COMDRWY2, RESDRWY2, DRWY1, DRWY2, LIGHT, VEI2, VCI2, GRADE1, GRADE2, HI1, PKTURN PKLEFT, PKLEFT1
L3LT
MEDTYPE3, PKLEFT1, PKTHRU2
 

Table 120 . Summary of Correlations for Independent Variables for Type IV Sites (Continued)

Variable
Positive Correlates1
Negative Correlates1
PKTRUCK
MEDTYPE2, L1RT, L1LT, L3RT, SPD1, PKTHRU2, SD1, SDL2, SDR2
AADT1, AADT2, HAZRAT1, DRWY1, LIGHT, TERRAIN1, PKTURN, PKLEFT, PKLEFT1,
PKTURN
AADT2, RESDRWY1, DRWY1, LIGHT, PKLEFT, PKLEFT1, PKLEFT2
AADT1, MEDTYPE1, MEDTYPE2, MEDWDTH1, L1RT, L1LT, L3RT, SPD1, PKTRUCK, PKTHRU1
VEI1
VI1, VCI1, GRADE1
MEDTYPE2, SD1, SDL2, SDR2
HEI1
AADT2, L1LT, RESDRWY1, DRWY1, DRWY2, LIGHT,
SHOUDLER2, L1LT, SPD1, SPD2, SD1, SDL2
GRADE1
MEDTYPE1, HAZRAT1, TERRAIN1, VEI1,VI2, HI1
AADT2, MEDTYPE2, L1RT, PKTHRU2, SD1, SDL2, SDR2
SDL2
MEDTYPE2, HAZRAT2, HAZRAT2,VCI2, SPD1, PKTRUCK, SD1, SDR2
HAZRAT1, COMDRWY1, RESDRWY1, DRWY1, TERRAIN1, VEI1, VCEI1, GRADE1, HI1, HEI1
SDR2
MEDTYPE2, SHOULDER1, LEGACC2, HAZRAT2, RESDRWY2, VCI2, SPD1, PKTRUCK, PKTHRU2, SD1, SDL2
HAZRAT1, COMDRWY1, RESDRWY1, DRWY1, TERRAIN1, VEI1, VCEI1, GRADE1, HI1

1 Only those correlations are shown for which p-values are less than 0.10.

3.2.6 Type V

A summary of the full data for Type V intersections is shown in table 121. In total, 53 variables were available for model development. The HSIS California data were again excluded because only five Type V sites were available. This left the California and Michigan sites from the original study, with the additional years of accident data for inclusion in the database. Some California sites experienced changes in some design features during 1996-98 period. For these, only 1993-95 data were used. As before, the frequency column indicates the number of sites for which the information was available.

Table 121. Summary Statistics for Type V Sites

Variables
Frequency
Mean
Median
Minimum
Maximum
TOTACC per YEAR
100
5.9
5.3
0.0
26.5
INJACC per YEAR
100
1.8
1.5
0.0
6.5
AADT1
100
9126
8700
430
25132
AADT2
100
3544
3100
420
12478

Table 121 . Summary Statistics for Type V Sites (Continued)

Variables
Frequency
Mean
Median
Minimum
Maximum
SIGTYPE Total
100
N/A1
0:Pre-timed
33(33%)
1:Actuated
45(45%)
2:Semi-actuated
22(22%)
MEDTYPE on major Total
100
N/A1
0:No Median
87(87%)
1:Painted
12(12%)
2:Other
1(1%)
MEDTYPE on minor Total
51
 
0:No Median
48(94.1%)
 
1:Painted
3(5.9%)
 
2:Other
0(0%)
N/A1
MEDWDTH1
100
1.3
0
0
13
MEDWDTH2
100
0.3
0
0
12
SHOULDER1
51
1.9
2
0
10
SHOULDER2
51
1.5
2
0
10
L1RT Total
100
N/A1
0
51(51%)
1
21(21%)
2
28(28%)
L1LT Total
100
N/A1
0
23(23%)
1
2(2%)
2
75(75%)
L3RT Total
100
N/A1
0
59(59%)
1
20(20%)
2
21(21%)
L3LT Total
100
N/A1
0
45(45%)
1
5(5%)
2
50(50%)
LEGACC1 Total
51
 
0
46(90.2%)
 
1
5(9.8%)
N/A1

1 N/A: not available

Table 121 . Summary Statistics for Type V Sites (Continued)

Variables
Frequency
Mean
Median
Minimum
Maximum
LEGACC2 Total
51
N/A1
0
50(98%)
1
1(2%)
PROTLT1 Total
100
N/A1
0
70(70%)
1
30(30%)
PROTLT2 Total
51
N/A1
0
47(92.2%)
1
4(7.8%)
HAZRAT1 Total
100
N/A1
1
12(12%)
2
29(29%)
3
27(27%)
4
16(16%)
5
13(13%)
6
3(3%)
7
0(0%)
HAZRAT2 Total
51
N/A1
1
1(2%)
2
8(15.7%)
3
17(33.3%)
4
4(27.5%)
5
8(15.7%)
6
3(5.9%)
7
0(0%)
COMDRWY1
100
2.64
2
0
11
RESDRWY1
100
0.52
0
0
6
DRWY1
100
3.16
3
0
15
COMDRWY2
100
2.44
2
0
10
RESDRWY2
100
0.69
0
0
8
DRWY2
100
3.13
3
0
11
LIGHT Total
100
N/A1
0
29(29%)
1
71(715)

1 N/A: not available

Table 121 . Summary Statistics for Type V Sites (Continued)

Variables
Frequency
Mean
Median
Minimum
Maximum
TERRAIN1 Total
100
N/A1
Flat
59(59%)
Rolling
38(38%)
Mountainous
3(3%)
TERRAIN2 Total
51
N/A1
Flat
18(35.3%)
Rolling
31(60.8%)
Mountainous
2(3.9%)
SD1
100
1314
1246
235
2000
SD2
100
1213
1091
224
2000
SDL1
100
774
673
122
2000
SDL2
100
910
750
142
2000
SDR1
51
822
798
103
2000
SDR2
51
1042
934
224
2000
VEI1
100
1.45
1.19
0.00
11.97
VEI2
100
1.91
1.39
0.00
13.50
VEICOM
100
1.81
1.59
0.00
8.13
VCEI1
100
1.10
0.45
0.00
10.79
VCEI2
100
1.54
0.90
0.00
14.00
VCEICOM
100
1.32
1.03
0.00
7.00
GRADE1
100
1.20
1.00
0.00
4.98
GRADE2
100
1.50
1.28
0.00
7.79
HEI
100
3.95
0.61
0.00
94.87
HI
100
2.15
0.00
0.00
60.00
HEI2
100
2.52
0.00
0.00
36.41
HI2
100
2.58
0.00
0.00
47.44
HEICOM
100
2.56
0.58
0.00
32.54
HICOM
100
2.36
0.00
0.00
42.05
HAU
100
0.07
0.00
-45.00
40.00
SPD1
100
45.2
45
25
65
SPD2
100
40.9
40
20
55
PKTRUK
49
8.96
7.71
2.69
45.43
PKTURN
49
35.64
34.48
7.07
72.66
PKTHRU1
49
71.19
73.77
18.01
96.73
PKTHRU2
49
43.90
41.99
8.45
84.09
PKLEFT
49
18.17
17.97
4.20
37.07
PKLEFT1
49
14.99
13.15
1.78
43.23
PKLEFT2
49
28.21
24.88
2.59
75.73

1 N/A: not available

Table 122 shows correlation statistics and p-values that indicate the association between crash counts and the independent variables for type V intersections. Table 123 shows correlations between the independent variables. Only those correlations that are significant at the 90 percent level are shown.

Again, as expected, major and minor road AADTs correlate positively with crashes. Peak turning movement volume also correlates with crashes, both positively and negatively. Shoulder width on major and minor roads, left-and right-lane on major and minor roads, acceleration lane on major and minor roads, protected left lane on major and minor roads, residential driveway on major and minor roads, terrains, sight distance, vertical curves, absolute grades, horizontal curves, intersection angle, and other variables are correlated with crashes in the opposite direction than expected, although many of these correlations are insignificant.

Table 122. Correlation Between Crashes and Independent Variables for Type V Sites

Variables
TOTACC per YEAR
INJACC per YEAR
Corr.
p-value
Corr.
p-value
AADT1
0.2581
0.0095
0.2964
0.0027
AADT2
0.4313
0.0000
0.3056
0.0020
MEDWDTH1
-0.0095
0.9251
0.0123
0.9035
MEDWDTH2
-0.0385
0.7036
-0.0942
0.3513
SHOULDER1
0.2324
0.1008
0.2826
0.0445
SHOULDER2
0.0818
0.5684
0.0557
0.6979
L1RT
0.2271
0.0231
0.1591
0.1138
L1LT
0.1516
0.1323
0.2033
0.0424
L3RT
0.2883
0.0036
0.2113
0.0348
L3LT
0.2178
0.0295
0.0771
0.4458
LEGACC1
0.3602
0.0094
0.2391
0.0911
LEGACC2
0.1079
0.4510
0.1461
0.3064
PROTLT1
0.1340
0.1837
0.1408
0.1622
PROTLT2
0.3652
0.0084
0.2452
0.0828
COMDRWY1
0.1012
0.3163
-0.1315
0.1922
RESDRWY1
-0.0130
0.8976
-0.0500
0.6212
DRWY1
0.0850
0.4004
-0.1377
0.1718
COMDRWY2
0.0015
0.9883
-0.1598
0.1122
RESDRWY2
-0.1924
0.0552
-0.0474
0.6399
DRWY2
-0.1149
0.2552
-0.1633
0.1044
LIGHT
-0.1885
0.0603
-0.2801
0.0048
SD1
0.1064
0.2919
0.1325
0.1888
SD2
0.1072
0.2886
0.1667
0.0975
SDL1
0.1692
0.0925
0.2437
0.0146
SDL2
0.1400
0.1649
0.2545
0.0106
SDR1
0.2057
0.1475
0.1938
0.1731
SDR2
0.0692
0.6296
0.1321
0.3556

Table 122 . Correlation Between Crashes and Independent Variables for Type V Sites (Continued)

Variables
TOTACC per YEAR
INJACC per YEAR
Corr.
p-value
Corr.
p-value
VEI1
0.1228
0.2234
0.0510
0.6144
VEI2
0.0378
0.7090
0.0467
0.6443
VEICOM
0.1276
0.2059
0.1032
0.3070
VCEI1
0.1167
0.2474
0.0229
0.8208
VCEI2
0.0376
0.7103
0.0275
0.7857
VCEICOM
0.1009
0.3179
0.0367
0.7169
GRADE1
-0.0487
0.6302
-0.1739
0.0836
GRADE2
-0.0312
0.7580
-0.1208
0.2312
HEI
-0.0181
0.8578
-0.0292
0.7734
HI
-0.1541
0.1258
-0.0822
0.4162
HEI2
-0.0369
0.7155
-0.1023
0.3112
HI2
0.0222
0.8268
-0.0070
0.9450
HEICOM
-0.1692
0.0924
-0.1403
0.1639
HICOM
-0.0882
0.3829
-0.0572
0.5722
HAU
-0.1326
0.1886
-0.1988
0.0474
SPD1
0.2103
0.0357
0.4325
0.0000
SPD2
0.1837
0.0674
0.3819
0.0001
PKTRUK
0.2097
0.1482
0.2116
0.1445
PKTURN
0.1950
0.1794
-0.1203
0.4105
PKTHRU1
-0.2396
0.0973
0.0702
0.6317
PKTHRU2
0.1079
0.4604
0.1468
0.3141
PKLEFT
0.2106
0.1464
-0.0904
0.5368
PKLEFT1
0.3471
0.0145
0.1895
0.1922
PKLEFT2
-0.2983
0.0374
-0.3784
0.0073

Table 123. Summary of Correlations for Independent Variables for Type V Sites

Variable
Positive Correlates1
Negative Correlates1
AADT1
AADT2, SIGTYPE2, MEDTYPE2, L1LT, PROTLT1, RESDRWY2, LIGHT, PKTHRU1
GRADE1, PKTURN, PKTHRU2, PKLEFT
AADT2
AADT1, SIGTYPE1, L1RT, L1LT, L3RT, L3LT, LEGACC1, SDR1, HEI1, HI2, PKTURN, PKLEFT, PKLEFT1
SIGTYPE3, HAZRAT1, HAZRAT2, GRADE2, HAU, PKTHRU1
PROTLT1
AADT1, SIGTYPE2, L1RT, L1LT, LEGACC1, LEGACC2, PROTLT2, RESDRWY2, TERRAIN2, VEI2,VEICOM, VCEI1, VCEICOM, HEI, HEI2, HI2, HEICOM, HICOM
SIGTYPE1, DRWY1, COMDRWY2,

Table 123 . Summary of Correlations for Independent Variables for Type V Sites (Continued)

Variable
Positive Correlates1
Negative Correlates1
MEDWDTH1
MEDTYPE1, MEDTYPE1minor, MEDWDTH2, L1LT, VEI2, VEICOM, VCEICOM, PKLEFT1
SIGTYPE1
HAU
TERRAIN1, VEI2, PKTHRU1
AADT2, METYPE1minor, MEDWDTH2, L3LT
HAZRAT1
SIGTYPE3, MEDTYPE2, HAZRAT2, TERRAIN1, VEI1, VCEI1, VCEICOM, GRADE1, GRADE2, HEICOM
AADT2, SIGTYPE1, SHOULDER1, SHOULDER2, L1RT, L1LT, L3RT, L4LT, LEGACC1, PROTLT2, SD1, SD2, SDL1, SDL2, SDR1, SDR2, SPD1, SPD2
DRWY1
COMDRWY1, RESDRWY1, COMDRWY2, DRWY2, LIGHT, PKTURN, PKLEFT, PKLEFT1
L1RT, L1LT, PRTLT1, SD2, SDL1, SDL2, SDR1, SPD1, SPD2, PKTHRU1
SPD1
SIGTYPE2, L1RT, L1LT, L3RT, L3LT, SD1, SD2, SDL1, SDL2, SDR1, SDR2, SPD2, PKTRUCK
HAZRAT1, HAZRAT2, COMDRWY1, DRWY1, COMDRWY2, DRWY2, LIGHT, VCEICOM, GRADE2, HEI1, PKTURN, PKLEFT
SPD2
L1RT, L1LT, L3RT, L3LT, SDD1, SD2, SDL1, SDL2, SDR1, SDR2, SPD1, PKTRUCK, PKTHRU2
HAZRAT1, HAZRAT2, COMDRWY1, DRWY1, COMDRWY2, RESDRWY2, DRWY2, LIGHT, GRADE2, HEI1, HEI2, HI2, HEICOM, HICOM, PKLEFT2
LIGHT (no=0,yes=1)
AADT1, SIGTYPE1, PROTLT1, COMDRWY1, DRWY1, COMDRWY2, DRWY2, PKLEFT2
SIGTYPE3, L1RT, L3RT, SDL1, SDL2, SPD1, SPD2, PKLEFT1
TERRAIN1
MEDTYPE2, HAZRAT1, TERRAIN2, VEI1, VEICOM, VCEI1, VCEICOM, GRADE1, GRADE2, HICOM, HAU
L1RT, L1LT, SD1, SD2, SDL1, SDL2, SDR2
L1RT
L1LT, L3RT, L3LT, LEGACC1, PROTLT1, SD1, SD2, SDL1, SDL2, SPD1, SPD2
HAZRAT1, COMDRWY1, DRWY1, LIGHT, TERRAIN1, VEI1, VCEI1, GRADE1, HEI1, HI1, HICOM
L1LT
AADT1, SIGTYPE2, MEDTYPE1, MEDWDTH1, L1RT, L3LT, PROTLT1, SD1, SDR2, SPD1, SPD2
SIGTYPE1, HAZRAT1, HAZRAT2, COMDRWY1, DRWY1, COMDRWY2, DRWY2, TERRAIN1, GRADE1

Table 123 . Summary of Correlations for Independent Variables for Type V Sites (Continued)

Variable
Positive Correlates1
Negative Correlates1
L3RT
AADT2, SHOULDER1, L1RT, L3LT, SDL1, SDL2, SDR1, SDR2, HEI2, HI2, SPD1, SPD2, PKTHRU2
HAZRAT1, DRWY2, LIGHT, VEI2, VEICOM, PKLEFT2
L3LT
AADT2, L1RT, L3RT, LEGACC1, PROTLT1, SD2, SDL2, VEI1, SPD1, SPD2, PKTHRU2
HAZRAT1, COMDRWY1, COMDRWY2, RESDRWY2, DRWY2, HAU, PKTHRU1
PKTRUCK
PROTLT1, SPD1, SPD2
 
PKTURN
AADT2, COMDRWY1, DRWY1, COMDRWY2, VEI1, VEICOM, VCEI1, VCEICOM, GRADE1, HEI1, HI2, PKLEFT, PKLEFT1
AADT1, SIGTYPE2, RESDRWY2, SPD1, PKTHRU1, PKTHRU2
VEICOM
MEDWDTH1, LEGACC2, PROTLT1, TERRAIN2, VEI1, VEI2, VCEI1, VCEI2, VCEICOM, GRADE1, GRADE2, HI1, PKTURN, PKLEFT1
L3RT, SD1, SDR1, SDR2, PKTHRU1
HEICOM
PROTLT1, HAZRAT1, HAZRAT2, VEI1, GRADE1, GRADE2, HEI1, HI1, HEI2, HI2, HICOM, PKLEFT2
SD1, SD2, SDL1, SDL2, SDR1, SPD2
GRADE1
HAZRAT1, HAZRAT2, TERRAIN1, VEI1, VEICOM, VCEI1, VCEICOM, GRADE2, HI1, HEI2, HEICOM, HICOM, PKTURN, PKLEFT
AADT1, L1RT, L1LT, SD1, SD2, SDL1, SDL2, SDR2, PKTHRU1
SDL2
SHOULDER1, L1RT, L3RT, L3LT, SD1, SD2, SDL1, SDR1, SDR2, SPD1, SPD2
HAZRAT1, HAZRAT2, COMDRWY1, DRWY1, DRWY2, LIGHT, TERRAIN1, TERRAIN2, GRADE1, GRADE2, HEI1, HI2, HEICOM, HICOM
SDR2
SHOULDER1, L1LT, L3RT, SD1, SD2, SDL1, SDL2, SDR1, SPD1, SDP2
HAZRAT1, HAZRAT2, TERRAIN1, VEI1, VEI2, VEICOM, VCEI1, VCEI2, VCEICOM, GRADE1, GRADE2

1 Variables only significant with p-value of 0.1 were selected

Previous | Table of Contents | Next

ResearchFHWA
FHWA
United States Department of Transportation - Federal Highway Administration