• 검색 결과가 없습니다.

Analysis on Comparison of Highway Accident Severity between Weekday and Weekend using Structural Equation Model

N/A
N/A
Protected

Academic year: 2021

Share "Analysis on Comparison of Highway Accident Severity between Weekday and Weekend using Structural Equation Model"

Copied!
10
0
0

로드 중.... (전체 텍스트 보기)

전체 글

(1)

** ᩑᖙݡ⦺Ʊ Ŗŝݡ⦺ ࠥ᜽Ŗ⦺ŝ ᕾᔍŝᱶ ([email protected])

Received June 21, 2013/ revised June 24, 2013/ accepted June 27, 2013

Copyright ⵑ 2013 by the Korean Society of Civil Engineers

 ǣŠ––’ǣȀȀ†šǤ†‘‹Ǥ‘”‰ȀͳͲǤͳʹ͸ͷʹȀ•…‡ǤʹͲͳ͵Ǥ͵͵Ǥ͸ǤʹͶͺ͵ ™™™Ǥ•…‡Œ‘—”ƒŽǤ‘”Ǥ”

ጪ⸮⇧ⷓ⢛#⁦㯓ⴂ#ⴲⱧ㬚#⺺⻏ኺ#⺺ᾎⴖ#ኞ❋ᦂḚ#♪ኞ⢪ᆿᦂ#␂ጎ⍂⛛

ࢼଗլ ȵੲটઽ ȵ୨஼෭

Bae, Yun Kyung*, Ahn Sunyoung**, Chung, Jin-hyuk***

Analysis on Comparison of Highway Accident Severity between Weekday and Weekend using Structural Equation Model

ABSTRACT

In order to identify and understand the crucial factors to induce traffic accident, causal relationships between diverse factors and traffic accident occurrence have been investigated continuously. It is one of most important issues all over the world to reduce the number of traffic accidents and deaths by them. Korea government is also stepping up their effort to reduce the number of traffic accidents and mitigate the severity of the accidents by establishing various traffic safety strategies. By introducing the five-day work week and increasing concern of leisure activities, the differences of trip characteristics between weekday and weekend is getting greater.

According to this, the patterns and crucial factors of traffic accident occurrence in weekend appear differently from those in weekday.

This study aims to understand major different factors affecting accident severity between weekday and weekend using 12,042 incident data occurred on freeways of Korea from 2006 to 2011. The model developed in this study estimated relationships among various exogenous factors of traffic accident by each type using SEM(Structural Equation Model). The result provides that road factors are related to the accident severity for weekday model, while environment factors affects on accident severity for weekend.

Key words : Traffic accident, Severity, SEM, Highway, Weekend

Ⅹಾ

Ʊ☖ᔍŁෝqᗭ᜽┅Łᦩᱥᖒᮥ⨆ᔢ᜽┅ʑ᭥⧕, ᔍŁᨱᩢ⨆ᮥၙ⊹۵᫵ᯙॅᮥᇥᕾ⦹ᩍƱ☖ᔍŁෝᩩ⊂⦹۵༉⩶ᮡḡᗮᱢᮝಽ}ၽࡹᨕ

᪵݅. Ʊ☖ᔍŁqᗭ᪡ᔍŁᔍ฾ᯱqᗭᨱݡ⦽ྙᱽ۵ᖁḥǎŝ⬥ḥǎ༉ࢱᵲ᫵⦽ྙᱽಽᕽ, ᬑญӹ௝ᨱᕽ۵ๅ⧕Ʊ☖ᔍŁqᗭෝ᭥⦽ᱶ

₦ၰႊᦩᮥᱽ᜽⦹Łᯩ݅. ᵝ5ᯝᱽ᜽⧪ŝᩍa⪽࠺ᨱݡ⦽šᝍ᮹᷾aಽᵝᵲŝᵝั᮹☖⧪➉▕ᮡ݅ෙ⩶┽ෝḡܩŁᯩᮝ໑, Ʊ☖ᔍŁ ᮹✚ᖒࠥษ₍aḡಽ݅෕íӹ┡ӽ݅. ᅙᩑǍᨱᕽ۵ᵝᵲŝᵝั᮹ᔍŁᝍbࠥᨱၙ⊹۵᫵ᯙᯕ݅෥ᨱݡ⦹ᩍᇥᕾ⦹ᩡᮝ໑, ᯕ۵Łᗮࠥ

ಽᨱᕽၽᔾ⦽5֥࠺ᦩ᮹ᔍŁᯱഭෝǍ᳑ႊᱶ᜾༉⩶ᮥᯕᬊ⦹ᩍᇥᕾ⦹ᩡ݅. ༉⩶᮹⇵ᱶđŝᵝᵲᔍŁᝍbࠥ༉⩶ᮡࠥಽ᫵ᯙᯕᔍŁᝍ bࠥᨱaᰆⓑᩢ⨆ᮥၙ⊹۵äᮝಽӹ┡ԍᮝ໑, ᵝัᔍŁᝍbࠥ༉⩶ᮡᬕᱥᯱ᫵ᯙᯕᔍŁᝍbࠥᨱaᰆⓑᩢ⨆ᮥၙ⊹۵äᮝಽӹ┡ӹ

ᵝᵲŝᵝัᔍŁ᮹ᬱᯙ₉ᯕෝӹ┡ԩ݅.

áᔪᨕ Ʊ☖ᔍŁ, ᝍbࠥ༉⩶, Ǎ᳑ႊᱶ᜾, Łᗮࠥಽ, ᵝั

”ƒ•’‘”–ƒ–‹‘‰‹‡‡”‹‰ İࣀėॡ

(2)

1. ᕽು

1.1 ઴֜ଭࢼլࢫࡧୡ

ᬑญӹ௝᮹Ʊ☖ᔍŁÕᙹ۵ๅ⧕qᗭ⦹۵⇵ᖙෝӹ┡ԕŁ

ᯩḡอ, ᵝ᫵ᖁḥǎᨱእ⧕ᕽ۵ᩍᱥ⯩׳ᮡᝅᱶᯕ݅. Ʊ☖ᔍŁ

qᗭ᪡ ᔍŁᔍ฾ᯱ qᗭᨱ ݡ⦽ྙᱽ۵ ᖁḥǎŝ ⬥ḥǎ༉ࢱ

ᵲ᫵⦽ྙᱽಽᕽ, ᬑญӹ௝ᨱᕽ۵ๅ⧕Ʊ☖ᔍŁqᗭෝ᭥⦽ᱶ₦

ၰႊᦩᮥᱽ᜽⦹Łᯩ݅. ᵝ5ᯝᱽ᜽⧪ŝᩍa⪽࠺ᨱݡ⦽šᝍ᮹

᷾aಽ ᵝᵲŝ ᵝั᮹ ☖⧪➉▕ᮡ ݅ෙ ⩶┽ෝ ḡܩŁ ᯩᮝ໑, Ʊ☖ᔍŁ᮹✚ᖒࠥษ₍aḡಽ݅෕íӹ┡ӽ݅. Ğₑℎ☖ĥᨱ

᮹⦹໕ 2011֥ ᬑญӹ௝ ࠥಽᨱᕽ ၽᔾ⦽ ᱥℕ ᔍŁ Õᙹ۵

221,711Õᯕ໑, ᵝัᯙ☁᫵ᯝŝᯝ᫵ᯝᨱၽᔾ⦽ᔍŁ۵62,638 Õᮝಽᯝᵝᯝ⠪Ɂᅕ݅ⓑᙹ⊹ෝӹ┡ԕ໑. ✚⯩ӹᔍ฾ᔍŁ۵

ᯝᵝᯝ⠪Ɂᯙ747Õᅕ݅׳ᮡᙹ⊹ෝӹ┡ԙ݅. ݡ⩶Ʊ☖ᔍŁ

☖ĥᨱᕽ۵ᵝᵲᔍŁෝ༉ࢱ⧊⊽ᱶࠥ᮹ᔍŁÕᙹaᵝัᨱၽᔾ

⦹ᩡᮝ໑, ᔍ฾ᔍŁ᮹Ğᬑᵝᵲၽᔾ⦽ᔍŁÕᙹᯙ57Õᅕ݅׳ᮡ

ᙹ⊹ᯙ71Õᯕᵝัᨱၽᔾ⦹ᩡ݅. ᯕ۵ᵝᵲŝᵝั᮹☖⧪➉▕

✚ᖒᯕ݅෕Ł, Ʊ☖ᔍŁᝍbࠥᨱၙ⊹۵᫵ᯙᯕ݅෕݅۵äᮥ

᮹ၙ⦹໑ᵝᵲŝᵝั᮹ᔍŁšญ᮹₉ᄥ⪵᮹⦥᫵ᖒᮥั⧕ᵝŁ

ᯩ݅. ੱ⦽ᵝั᮹☖⧪ᮡᵝᵲ☖⧪ŝ݅෕íݡᇡᇥᩍa☖⧪᮹

ᖒĊᯕv⦹ʑভྙᨱᬕᱥᯱ᮹⧪┽ӹᯖᙺ⦹ḡᦫᮡࠥಽᔢ⫊

॒ᨱ ঑ෙ ᔍŁᬱᯙᯕ ⓕ äᮝಽ ᔾbࡽ݅.

ᅙᩑǍᨱᕽ۵ŁᗮࠥಽᔍŁᯱഭෝᯕᬊ⦹ᩍᵝᵲŝᵝั᮹

ᔍŁᝍbࠥᨱၙ⊹۵ᩢ⨆ᨱݡ⦹ᩍᇥᕾ⦹ᩡ݅. ᔍŁÕᙹ᪡⧉̹

ᔍŁ᮹ᝍbࠥ۵ᵲ᫵⦹íᯙ᜾ࡹ۵ḡ⢽ᯕ໑, ࠥಽ᮹ʑ⦹Ǎ᳑ӹ

ᬕᱥᯱ⧪┽, ₉᳦, ԁᦉ॒ฯᮡ᫵ᯙॅᨱ᮹⧕Ḣ,eᱲᱢᯙᩢ⨆ᮥ

ၼᮥäᮝಽ❱݉ࡽ݅. ᯕ్⦽᫵ᯙॅᯕᅖᰂ⦽šĥෝaḡ໕ᕽ

ᔍŁ᮹ᝍbࠥᨱᩢ⨆ᮥၙ⋁äᮝಽᅕᯕ໑, Ǎ᳑ႊᱶ᜾༉⩶ᮡ

ᯕ్⦽ᄡᙹॅ᮹šĥෝ Ƚ໦⦹۵ߑᱢ⧊⦽༉⩶ᮝಽ❱݉ࡹᨕ

ᅙ ᩑǍᨱᕽ ᔍᬊ⦹ᩡ݅.

1.2 ઴֜ଭ࣐଍ࢫࢺ࣑

ᅙᩑǍᨱᕽ۵2006֥ᇡ░2010֥ʭḡŁᗮࠥಽᨱᕽၽᔾ⦽

ᔍŁᯱഭෝᯕᬊ⦹ᩍᵝᵲŝᵝั᮹ᔍŁᝍbࠥᨱၙ⊹۵ᩢ⨆ᨱ

ݡ⦹ᩍᇥᕾ⦹ᩡ݅. Łᗮࠥಽᨱᕽၽᔾ⦽ᯱഭ۵š⧁ḡᔍ, ᯝᯱ,

᜽e, יᖁ, ᭥⊹, ႊ⧪॒᮹ᔍŁᨱš⦽Ǎℕᱢᔍ⧎ᮥʑಾ⦹ᩍ

šญ⦹Łᯩ݅. ᵝᵲ᮹ᔍŁÕᙹ۵8,152Õ, ᵝั᮹ᔍŁÕᙹ۵

3,530ÕᮝಽḲĥࡹᨩᮝ໑ᱥℕᔍŁÕᙹ۵12,042Õᯕ݅. ᇥᕾݡ ᔢŁᗮࠥಽ۵Ğᇡᖁ, Ğᯙᖁ, ᱽ2Ğᯙᖁ, ᩢ࠺ᖁ, ᕽᬙ᫙Şᙽ⪹ᖁ, ᵲᇡᖁ, ᱽ2ᵲᇡᖁ, ᕽ⧕ᦩᖁ॒ⅾ24}ᱥǎ᮹Łᗮࠥಽᯕ݅.

༉⩶ᮡǍ᳑ႊᱶ᜾ᮥᱢᬊ⦹ᩍ⇵ᱶ⦹ᩡᮝ໑, AMOS 20 ⥥ಽəఉ

ᮥᯕᬊ⦹ᩡ݅. ᔍŁ᮹ᝍbࠥᨱၙ⊹۵᫵ᯙᮡ݅ෙᔍŁǍ᳑ႊᱶ

᜾༉⩶ᨱᕽᔍᬊ⧕᪉ä⃹ౝࠥಽ᫵ᯙ, ᬕᱥᯱ᫵ᯙ, ⪹Ğ᫵ᯙᮝಽ

Ǎᇥ⦹ᩍ b ᫵ᯙॅᮥ ݡ⢽⧁ ᙹ ᯩ۵ ᄡᙹॅᮥ ᖁᱶ⦹ᩡ݅.

2. šಉᩑǍŁₑ

Ʊ☖ᔍŁᨱš⦽ᩑǍ۵᪅௹ᱥᇡ░ḡᗮᱢᮝಽᯕ൉ᨕᲙ᪵ᮝ ໑, əᬱᯙᮥᇥᕾ⦹ᩍ}ᖁ⦽Łšญ⦹ʑ᭥⦽יಆᮡ⪽ၽ⦹í

ᯕ൉ᨕᲭ݅. ݅᧲⦽ᇥᕾႊჶŝᯱഭෝᯕᬊ⦹ᩍᔍŁෝᩩႊ⦹ʑ

᭥⦽ᩑǍॅᯕḥ⧪ࡹᨩ݅. ǎ☁ᩑǍᬱ(1996)᮹ᩑǍᨱᕽ۵Ʊ☖

ᦩᱥŝšಉᯕᯩ۵ࠥಽʑ⦹Ǎ᳑᫵ᗭॅᮥࠥ⇽⦹ᩡ݅. ₉ಽ⡎, łᖁၹĞ, ᖁ⩶᮹ᩑᗮᖒ, ᳦݉Ğᔍ, ᜽Ñ, ʙᨕˉ⡎, ᵲᦺᇥญݡ᮹

7aḡࠥಽʑ⦹Ǎ᳑᫵ᯙᮥƱ☖ᔍŁ᮹ᵝ᫵ᩢ⨆᫵ᯙᮝಽᖁᱶ⦹

ᩡ݅. ᯕ7aḡ᫵ᯙᨱݡ⦹ᩍᔢݡᱢᯙᔍŁ᭥⨹ࠥෝӹ┡ԝᙹ

ᯩ۵ᔍŁ᭥⨹ࠥ❱݉ḡᙹෝ᳦ᗮᄡᙹಽ₥┾⦹ᩍ݉ᙽ⫭ȡ༉⩶ᮥ

⇵ᱶ⦹ᩡ݅. ၶ⬉ᝁ(2007)ᮡ Ʊ☖ప, ࠥಽ ʑ⦹Ǎ᳑ ॒ŝ zᮡ

Ʊ☖ᔍŁ᫵ᯙॅŝƱ☖ᔍŁ᪡᮹šĥෝᇥᕾ⦹ᩡ݅. ᇥᕾ᮹ݡᔢ ḡᩎᮡŁᗮࠥಽ᮹✙ౝ⠌ᯙ░ℕᯙḡෝᇥᕾݡᔢᮝಽ⦹ᩡᮝ໑

ᔍŁ༉⩶ᮡᮭᯕ⧎ᇥ⡍ෝᱢᬊ⦹ᩍᩑđಽ᮹⩶┽ᄥಽᮭᯕ⧎⫭

ȡ༉⩶ᮥ⇵ᱶ⦹ᩡ݅. ༉⩶ᨱᯕᬊࡽᄡᙹ۵Ʊ☖ప, ᵲ₉పእᮉ,

ࠥಽ᮹łශၰłᖁᰆ, ᮁ⇽᯦ᩍᇡ॒ᯕ໑ᯕ్⦽ᄡᙹॅŝᔍŁ۵

šಉᯕ ᯩᮭᮥ ᅕᩡ݅.

ᯕᵝᩑ(2008)ᮡǍ᳑ႊᱶ᜾༉⩶ᮥᯕᬊ⦹ᩍŁᗮࠥಽ᮹Ʊ☖ᔍ Łᝍbࠥᨱᩢ⨆ᮥၙ⊹۵᫵ᯙॅᮥᇥᕾ⦹ᩡ݅. ᔍŁᝍbࠥෝ

Ǎ᳑༉⩶᮹ԕᔾᰁᰍᄡᙹಽᖁᱶ⦹Ł, ࠥಽ᫵ᯙၰᬕᱥᯱ᫵ᯙ,

⪹Ğ᫵ᯙᮥ᫙ᔾᰁᰍᄡᙹಽᖅᱶ⦹ᩍ༉⩶ᮥ}ၽ⦹ᩡ݅. ⇵ᱶ

đŝ, ⠪໕ᖁ⩶ŝ ᳦݉Ǎ႑, י໕ᔢ┽ ၰ ԁᦉ۵ ᔍŁ ᝍbࠥ᪡

ᮭ᮹šĥᨱᯩ۵äᮝಽӹ┡ԍᮝ໑, ᵝ᧝᪡ᬱᯙ₉₉᳦, ᬕᱥᯱ

ᩑಚݡ۵ᔍŁᝍbࠥ᪡᧲᮹šĥᨱᯩ۵äᮝಽӹ┡ԍ݅. ʡᔢಾ

(2010) ᮡ݉ᗮඹ᜽ᖅᨱᕽၽᔾ⦽Ʊ☖ᔍŁᮁ⩶ᮥ₉ݡ₉ᔍŁ᪡

₉ݡ ᔍ௭ᔍŁಽ Ǎᇥ⦹ᩍᔍŁ᮹ ᵝ᫵ᯙᮥ Ǎ᳑ႊᱶ᜾ ༉⩶ᮥ

ᯕᬊ⦹ᩍᇥᕾ⦹ᩡ݅. ᕽᬙ᜽ᕽݡྙǍᨱᕽၽᔾ⦽Ʊ☖ᔍŁᯱഭෝ

ᇥᕾ⦹ᩡᮝ໑Ʊ☖ᔍŁᝍbࠥ᪡᫙ᔾᱢᄡᙹॅe᮹šĥෝ⇵ᱶ⦹

ᩡ݅. ₉ݡ₉ᔍŁᨱᕽ۵ࠥಽ᫵ᯙᯕᩢ⨆ᮥၙ⊹Ł, ₉ݡᔍ௭

ᔍŁᨱᕽ۵⪹Ğᱢ᫵ᯙᯕⓍíᩢ⨆ᮥၙ⊹۵äᮝಽᇥᕾ⦹ᩡ݅.

✚⯩ӹ ᵝัŝ ᵝᵲ᮹ ᔍŁᨱ š⦽ ᩑǍॅŝ Ŗ⮕ᯝ ᔍŁᨱ

š⦽ᩑǍॅࠥ݅᧲⦹íḥ⧪ࡹᨩ݅. ᯕ్⦽ᩑǍॅᮡᵝᵲŝᵝั

᮹Ʊ☖➉▕ᯕ݅෕Ł, ᵝั᮹Ʊ☖ᔍŁ᮹ᝍbࠥaᵝᵲ᮹ᔍŁᅕ

݅ᝍb⦹݅Łᱽ᜽⦹Łᯩ݅(Gray et al., 2008; Quddus et al., 2010; Barua & Tay, 2010).

Gray et al.(2008)ᮡᩢǎ᮹ࠥಽᦩᱥࠥ⊂ᱶᮥ᭥⦹ᩍ13֥e ᮹ᯱഭෝᯕᬊ⦹ᩍᔍŁ᮹ᝍbࠥෝᇥᕾ⦹ᩡ݅. ᙽᕽ⩶⥥ಽኸ

༉⩶ᮥᯕᬊ⦹ᩡᮝ໑, ᔍᬊࡽᄡᙹ۵ᔍŁ᭥⊹, ᭥ၹᩍᇡ, ᩑಚ,

(3)

Table 1. Elements of Structural Equation Model

x q×1 vector of observed exogenous variables

Measurement Model

y p×1 vector of observed endogenous variables 쩍 n×1 vector of latent exogenous variables 쩆 m×1 vector of latent endogenous variables

쩃 q×1 vector of measurement error terms for observed variables x 쩄 p×1 vector of measurement error terms for observed variables y

쨪x The matrix(q×n) of structural coefficients for latent exogenous variables n to their observed indicator variables q 쨪y The matrix(p×m) of structural coefficients for latent exogenous variables m to their observed indicator variables p Structural Model

쨢 The matrix(m×n) of regression effects for latent exogenous variables n to latent endogenous variables m 쨡 The coefficient matrix (m×m) of direct effects between latent endogenous variables

Fig. 1. Structural Equation Model ḡᩎᱢ ⬉ŝ, ĥᱩᱢ ⬉ŝ, ᵝᵲ/ᵝั, ʑᔢᔢ┽, ࠥಽ ⢽໕ᔢ┽

॒᮹ᩍ్aḡ⬉ŝෝᇥᕾ⦹ᩡ݅. ᵝᵲ, ᵝe, ᱡᗮࠥᨱᕽၽᔾ⦹

۵ᔍŁ۵ᔍŁ᮹ᝍbࠥaԏíӹ┡ӽၹ໕, ᵝั, ᧝e, Łᗮ₉ప, ᔍŁ݅ၽḡᱱ॒ᨱᕽၽᔾ⦹۵ᔍŁ۵⊹໦ᱢᯙᔍŁaฯᯕၽᔾ

⦹۵äᮝಽᇥᕾ⦹ᩡ݅. Barua and Tay(2010)ᮡႊɡ௝ߑ᜽᪡

zᮡ}ၽࠥᔢǎᨱᕽၽᔾ⦹۵ݡᵲƱ☖ᔍŁෝᙽᕽ⩶⥥ಽኸ༉

⩶ᮝಽᇥᕾ⦹ᩡ݅. 1998֥ᇡ░2005֥ʭḡ᮹ᔍŁᯱഭෝᯕᬊ⦹

ᩡᮝ໑, ᔍŁÕᙹ᪡ ᔍŁᮉᮡ ᜽eᯕ ḡԉᨱ ঑௝ ᱱ₉ ᝍb⦽

ᔢ┽ᩡ݅. ᔍŁ᮹ᝍbࠥ۵ᵝั, እ℉ࢱ, ᧲ႊ⨆₉ಽ᮹ᔢ⫊ᨱᕽ

޵ ᝍb⦽ äᮝಽ ӹ┡ԍ݅.

ੱ⦽Anowar et al.(2013)᮹ᩑǍᨱᕽ۵ᵝัŝǎaŖ⮕ᯝᨱ

ၽᔾ⦹۵ᔍŁෝᩩႊ⦹Ł, ᱶ₦ᮥḲ⧪⦹ʑ᭥⦹ᩍᔍŁᨱᩢ⨆ᮥ

ᵝ۵ ᫵ᯙᨱ ݡ⦹ᩍ ᇥᕾ⦹ᩡ݅. ⋱ӹ݅ᨱᕽ 5֥ ࠺ᦩ ၽᔾ⦽

Ŗ⮕ᯝŝᵝั᮹ᯱഭෝᯕᬊ⦹ᩡᮝ໑, ಽḡᜅ❒⫭ȡ༉⩶ᮥᔍᬊ

⦹ᩍᇥᕾ⦹ᩡ݅. ᵝᵲ᮹ᔍŁᅕ݅ᵝัŝŖ⮕ᯝᨱᔍŁa޵

ฯᯕၽᔾ⦹Ł, ᔍŁ᮹ᝍbࠥࠥ׳ᮡäᮝಽӹ┡ԍ݅. Ŗ⮕ᯝŝ

ᵝัᨱ᭥⨹⦽ᬕᱥ⧪┽a޵ฯᯕӹ┡ӹŁᗮࠥ᭥ၹࠥⓑäᮝಽ

ӹ┡ԍ݅.

3. Ǎ᳑ႊᱶ᜾༉⩶

Ǎ᳑ႊᱶ᜾༉⩶ᮡᔍ⫭⦺ၰᝍญ⦺ᨱᕽ}ၽࡽ⊂ᱶᯕುᨱ

☁ݡෝࢵ⪶ᯙᱢ᫵ᯙᇥᕾŝĥపĞᱽ⦺ᨱᕽ}ၽࡽᩑพႊᱶ᜾

༉ߙᨱ☁ݡෝࢵ݅ᵲ⫭ȡᇥᕾၰĞಽᇥᕾ॒ᯕđ⧊ࡽᖒĊᮥ

w۵ႊჶುᯕ௝Ł⧁ᙹᯩ݅(႑ᄲ಍, 2005, 2011). Ǎ᳑ႊᱶ᜾

༉⩶ᮡᯝၹᱢᮝಽ⊂ᱶ༉⩶ŝǍ᳑༉⩶ᮝಽǍᖒࡹᨕᯩ݅. ⊂ᱶ

༉⩶(Measurement model)ᮡbb᮹š⊂ᄡᙹa✚ᱶᰁᰍᄡᙹᨱ

ݡ⧕ᨕਜíᱢᰍࡹᨕᯩ۵aෝᖅᱶ⦽⪶ᯙᱢ᫵ᯙᇥᕾ᮹ᖒĊᮥ

wŁᯩᨕᕽ, š⊂ᄡᙹॅ᮹⊂ᱶᗮᖒᯙᝁ഑ࠥၰ┡ݚࠥෝ⠪a⧁

ᙹ ᯩ݅. Ǎ᳑༉⩶(Structural model)ᮡ ᨕਅ ᰁᰍᄡᙹa ݅ෙ

ᰁᰍᄡᙹ᮹ᄡ⪵ᨱḢᱲᱢᮝಽੱ۵eᱲᱢᮝಽᩢ⨆ᮥᵝŁᯩ۵ aෝᖅᱶ⦽݅ᵲ⫭ȡᇥᕾၰĞಽᇥᕾ᮹ᖒĊᯕၹᩢࡹᨕᰁᰍᄡ ᙹॅe᮹šĥෝᖅᱶ⦽äᯕ݅. Ǎ᳑ႊᱶ᜾ᮥ⇵ᱶ⦹۵⥥ಽəఉ ᨱ۵LISREL, AMOS, EQS ॒ᯕᯩᮝӹ, ᅙᩑǍᨱᕽ۵AMOS

ෝ ᯕᬊ⦹ᩍ ༉⩶ᮥ ⇵ᱶ⦹ᩡ݅.

Ǎ᳑ႊᱶ᜾༉⩶ᇥᕾ᮹ʑᅙᱢᯙŝᱶᮡ⊂ᱶᄡᙹෝ☖⦽ᰁᰍ

᫵ᯙᮥၽč⦹Łᰁᰍ᫵ᯙeᨱᯙŝšĥ᮹aᖅᮥᖅᱶ⦹۵äᯕ

݅. ᯝၹᱢᯙǍ᳑ႊᱶ᜾᮹Ǎᖒ᫵ᗭ۵݅ᮭTable 1ŝzᮝ໑, Fig. 1 ᮡ Ǎᖒ᫵ᗭॅᮥ Ğಽࠥ⩶ᮝಽ ӹ┡ԙ äᯕ݅.

AmosᨱᕽǍ᳑ႊᱶ᜾༉⩶᮹༉ᙹෝ⇵ᱶ⦹۵ʑჶᨱ۵↽ݡ ᬑࠥ⇵ᱶჶ(maximum likelihood), ᯝၹ⪵↽ᗭᯱ᜚ჶ(generalized least square), እaᵲ↽ᗭᯱ᜚ჶ(unweighted least square), ⃺ࠥ

ᯱᮁ↽ᗭᯱ᜚ჶ(scale-free least square) ॒ᯕᯩᮝ໑, ᅙᩑǍᨱᕽ ۵↽ݡᬑࠥ⇵ᱶჶᮥᯕᬊ⦹ᩡ݅. ↽ݡᬑࠥ⇵ᱶჶᨱᯕᬊࡹ۵༊ᱢ

⧉ᙹ۵݅ᮭŝz݅. ↽ݡᬑࠥ⇵ᱶჶ᮹ᱢ⧊⧉ᙹ۵݅ᮭŝz݅.

Ÿ

¦¥

á “–ŽÞľßâƒƐã¬

àÎ

Þľßäà “–Ž¬à ÞƎ âƏß (1)

ᯝၹᱢᮝಽaᰆฯᯕᯕᬊࡹ۵⇵ᱶჶᮡ↽ݡᬑࠥ⇵ᱶჶᮝಽ

ᯕ۵⇵ᱶపᯕݡȽ༉⢽ᅙᨱᕽᯝšᱢ(consistent)ᯕŁ, ᱱɝ⬉ᮉ

(4)

Table 2. Descriptive Statistics of Accident Records

Category Weekday Weekend

Frequency (person) Ratio (%) Frequency (person) Ratio (%)

Total 8,152 100.0 3,530 100.0

Parking on shoulder Relativeness 217 2.7 80 2.3

No relativeness 7,935 97.3 3,450 97.7

Horizontal alignment Straight 5,027 61.7 2,180 61.8

Curve 3,125 38.3 1,350 38.2

Vertical alignment Level 7,651 93.9 3,321 94.1

Slope 501 6.1 209 5.9

Accident location Tunnel 183 2.2 83 2.4

Others 7,969 97.8 3,447 97.6

Road surface Asphalt 3,583 44.0 1,535 43.5

Concrete 4,569 56.0 1,995 56.5

Weather condition Clear 4,634 56.8 1,931 54.7

Others(rain, fog, etc) 3,518 43.2 1,599 45.3

Daytime or nighttime Daytime 4,868 59.7 2,114 59.9

Nighttime 3,284 40.3 1,416 40.1

Surface condition Dry 5,905 72.4 2,520 71.4

Wet 2,247 27.6 1,010 28.6

Driver’s gender Female 824 10.1 516 14.6

Male 7,328 89.9 3,014 85.4

Driver’s age

20s than less 1,313 16.1 744 21.1

30~39 2,203 27.0 956 27.1

40~49 2,688 33.0 1,068 30.3

50~59 1,554 19.1 624 17.7

60s than more 394 4.8 138 3.9

High operating speed Normal 6,384 78.3 2,686 76.1

Speeding 1,768 21.7 844 23.9

Vehicle type small and medium 5,235 64.2 2,804 79.4

Full-size 2,917 35.8 726 20.6

Note: Vehicle type for small and medium size included cars(business, private, jeep) and full-size included freight, trailer and special cargo.

ᱢ(asymptotically efficient)ᯙäᮝಽӹ┡ӽ݅. ↽ݡᬑࠥ⇵ᱶჶ

ᱢ⧊⧉ᙹ۵⪶ශၡࠥ⧉ᙹෝᱢᬊ⦹ᩍᵝᨕḥ⢽ᅙᯕ⣡⩥⧁⪶ශ ၡࠥaaᰆ׳ᦥḡࠥಾ༉ᙹෝ⇵ᱶ⦽݅. ↽ݡᬑࠥ⇵ᱶჶᯕaᰆ

ฯᯕᔍᬊࡹ۵ᯕᮁ۵ ✚ᱶ⦽᳑Õ⦹ᨱᕽᱢ⧊⧉ᙹ᮹↽ᗭ⊹ᨱ

N-1 ᮥ Œ⦹໕ Ō

Ï

ᇥ⡍ෝ ⦹ʑ ভྙᯕ݅.

4. ༉⩶Ǎ⇶

4.1 ୀ߹ଭ׆ొധծंজ

ᅙᩑǍᨱᕽ۵2006֥ᇡ░2010֥ʭḡ᮹5֥࠺ᦩ᮹⦽ǎࠥಽ Ŗᔍᨱᕽš⧁⦹Łᯩ۵ĞᇡŁᗮࠥಽ, ᩢ࠺Łᗮࠥಽ, 88᪍ฝ⦞

Łᗮࠥಽ॒ᱥǎŁᗮࠥಽᨱᕽၽᔾ⦽Ʊ☖ᔍŁᯱഭෝᯕᬊ⦹ᩡ݅.

⧕ݚᯱഭ۵ᔍŁၽᔾḡ, ၽᔾ᜽e, ႊ⨆, ᔍŁ₉᳦, ᬕᱥᯱ᪡⦝⧕

ᯱ᮹ᖒᄥŝӹᯕ, ࠥಽᖁ⩶, ʑᔢᔢ┽, י໕ᔢ┽, ᔍŁ}᫵, ᔍŁᬱ ᯙ, ᔍ฾ᯱᙹ, ᵲᔢᯱᙹ, Ğᔢᯱᙹ॒݅᧲⦽ᱶᅕෝݕŁᯩ݅.

ᯕᵲᨱᕽᔢššĥa׳ᦥ݅ᵲŖᖁᖒ᮹ᬑಅaᯩ۵ᄡᙹॅᮥ

ᱽ᫙⦹Ł, đ⊂ᯱഭॅᮥݡℕੱ۵ᱽÑ⦹ᩡ݅. ᯕ్⦽ᯱഭ᮹

ᱱáᮥ☖⧕⇵⇽⧕ԙ16}ᄡᙹ(tʙᵝᱶ₉, ⠪໕ᖁ⩶, ᳦݉Ǎ႑, ᔍŁǍe, ࠥಽ⡍ᰆ, ᵝ᧝, ʑᔢᔢ┽, ࠥಽ⢽໕ᔢ┽, ᬕᱥᯱӹᯕ, ᬕᱥᯱᖒᄥ, ŝᗮᩍᇡ, ₉᳦, ᔍ฾ᯱᙹ, ᇡᔢᯱᙹ, ᔍŁ₉పᙹ,

₉ప❭ᗱݡᙹ)᮹12,042}ᯱഭෝᔍŁ᮹ᮁ⩶ᨱ঑௝ᵝᵲၽᔾ⦽

ᔍŁ᪡ᵝัᨱၽᔾ⦽ᔍŁಽᇥඹ⦹ᩍ ᯕᬊ⦹ᩡ݅. Table 2۵

(5)

Table 3. Definition of Model Variables

Category 1 0

Road factors

Parking on shoulder Relativeness No relativeness

Horizontal alignment Straight Curve

Vertical alignment Slope(-3%~3%) Slope(>3%)

Accident location Tunnel Others

Road surface Concrete Asphalt

Environment factors

Weather condition Clear Others(rain, fog, etc)

Daytime or nighttime Nighttime Daytime

Surface condition Dry Wet

Driver factors

Driver’s gender Male Female

Driver’s age 20s than less: 1, 30s: 2, 40s: 3, 50s: 4, 60s: 5

High operating speed Speeding Others

Vehicle type Full-size small and medium

⧕ݚᯱഭ᮹ʑᚁ☖ĥపᮥ᫵᧞⦽äᯕ݅. ᔍŁᯱഭ᮹ʑᚁ☖ĥ

ᇥᕾđŝෝᔕ⠕ᅕ໕, ᵝัŝᵝᵲᨱၽᔾ⦹۵ᔍŁ᮹Ǎᖒእᮉᯕ

₉ᯕaᯩᮭᮥ᦭ᙹᯩ݅. ᩩෝॅᨕᵝัŝᵝᵲᨱၽᔾ⦽ᔍŁ᮹

ᖒᄥ, ᩑಚ, ₉᳦ᇥඹ॒᮹Ğᬑ, əǍᖒእᮉᨱᕽ݅ෙᄡᙹ᪡

እƱ⦹ᩍ ₉ᯕa ᯩᮭᮥ ӹ┡ԙ݅. ᅙ ᩑǍᨱᕽ ᇥᕾᨱ ᔍᬊࡽ

༉⩶ᯙǍ᳑ႊᱶ᜾༉⩶ᮡᵝᵲŝᵝั᮹ᔍŁᝍbࠥᨱၙ⊹۵

ᩢ⨆ᮥእƱ⦹ʑ᭥⦽༉⩶ᮝಽᔍᬊࡹᨩ݅. Ǎ᳑ႊᱶ᜾༉⩶ᮡ

ᄡᙹॅe᮹ᅖᰂ⦽ᯙŝšĥෝ⧕ᕾ⦹Ł, ԕᔾᱢ, ᫙ᔾᱢᄡᙹॅe ᮹šĥෝ݅൉۵ߑᯩᨕᰆᱱᮥaḡŁᯩ݅. ᔍŁᝍbࠥᨱၙ⊹۵

ᩍ్ aḡ ᄡᙹॅᨱ ݡ⦹ᩍ ᄡᙹॅe᮹ šĥ᪡ ᄡᙹॅŝ ᔍŁ

ᝍbࠥ᪡᮹ šĥෝ ⧕ᕾ⦹۵ߑ ᰆᱱᮥ aḥ݅.

ᅙᩑǍᨱᕽ۵ᔍŁߑᯕ░᮹š⊂ᄡᙹॅᮥⓍíᔍŁᝍbࠥ᪡

ࠥಽ᫵ᯙ, ᬕᱥᯱ᫵ᯙ, ⪹Ğ᫵ᯙ॒ᮝಽᇥඹ⦹Ł, ᯕෝᰁᰍᄡᙹ ಽᖅᱶ⦹ᩍǍ᳑༉⩶ᮥ⇵ᱶ⦹ᩡ݅. ࠥಽ᫵ᯙᮡᬕᱥᯱ᮹✚ᖒᯕ ӹᵝᄡ⪹Ğᨱᩢ⨆ᮥၼḡᦫŁࠥಽ᮹Ǎ᳑ᱢ᫵ᗭᨱᩢ⨆ᮥ

ၙ⊹۵ᰁᰍᄡᙹಽᕽtʙᵝᱶ₉, ⠪໕ᖁ⩶, ᳦݉Ǎ႑, ᔍŁǍe,

ࠥಽ⡍ᰆᮥ š⊂ᄡᙹಽ ᖅᱶ⦹ᩡ݅. ⪹Ğ ᫵ᯙᮡ Ʊ☖ᔢ⫊ᯕӹ

⪹Ğ ॒ ࠥಽ ᫙ᱢᯙ ᄡ⪵ෝ ☖⦹ᩍ Ʊ☖᜽ᖅᯕӹ Ʊ☖⮱෥ᨱ

ᩢ⨆ᮥၙ⊹۵ᰁᰍᄡᙹಽᕽԁᦉ, ᵝ᧝eǍᇥ, ᔍŁၽᔾ᜽eᮥ

š⊂ᄡᙹಽᖅᱶ⦹ᩡ݅. ੱ⦽ᬕᱥᯱ᫵ᯙᮡᬕᱥᯱ᮹⧪┽ᨱᩢ⨆

ᮥၙ⊹۵᫵ᯙᮝಽᬕᱥᯱ᮹ᖒᄥŝᩑಚ, ŝᗮᩍᇡ, ᬱᯙ₉᳦ᮥ

š⊂ᄡᙹಽᖅᱶ⦹ᩡ݅. ᔍŁᝍbࠥᰁᰍᄡᙹෝӹ┡ԕ۵š⊂ᄡ ᙹಽ۵ᔍ฾ᯱᙹ, ᇡᔢᯱᙹ, ᔍŁ₉పᙹ, ₉ప❭ᗱݡᙹ॒4}ᄡᙹ

ෝᯕᬊ⦹ᩡ݅. 4}᮹ᰁᰍᄡᙹaᬕᨱᔍŁᝍbࠥ۵ԕᔾᄡᙹಽ,

ࠥಽ᫵ᯙŝᬕᱥᯱ᫵ᯙ, ⪹Ğ᫵ᯙ॒3}ᄡᙹ۵᫙ᔾᄡᙹಽ

ᖅᱶ⦹ᩍ ༉⩶ᮥ Ǎ⇶⦹ࠥಾ ⦽݅. ༉⩶ ⇵ᱶᮥ ᭥⦹ᩍ ᱢᬊࡽ

ᄡᙹॅᮡ Table 3ᨱᕽ ӹ┡ӽ݅.

4.2 ࡦ෴ଭ౟୨ࢫէր

Ǎ᳑ႊᱶ᜾ᨱ᮹⦽ᵝᵲᵝัᔍŁᝍbࠥ༉⩶ᮡAMOS(Analysis of Moment Structure)ෝᯕᬊ⦹ᩍǍ⇶⦹ᩡᮝ໑, ↽᳦ᱢᮝಽ⇵ᱶ

ࡽ ༉⩶ᮡ Fig. 2᪡ Fig. 3ŝ z݅.

ʑⅩ☖ĥᇥᕾ᮹đŝ᪡ʑ᳕ᩑǍॅ᮹đŝॅᮥ☁ݡಽᅙᩑǍ ᨱᕽ۵↽᳦ᱢᮝಽࠥಽ᫵ᯙŝ⪹Ğ᫵ᯙ, ᬕᱥᯱ᫵ᯙᮥ᫙ᔾ

ᰁᰍᄡᙹಽ, ᔍŁᝍbࠥෝԕᔾᰁᰍᄡᙹಽᖅᱶ⦹ᩍǍ᳑ႊᱶ᜾

༉⩶ᮥǍ⇶⦹ᩡ݅. ⇵ᱶࡽ༉ᙹॅᮡ༉ࢱ⢽ᵡ⪵⧕(standardized solution)ᯕ໑, ᯕෝ☖⧕᫙ᔾᰁᰍᄡᙹॅᯕԕᔾᰁᰍᄡᙹᨱၙ⊹

۵ᔢݡᱢᩢ⨆ಆᮥ❭ᦦ⧁ᙹᯩ݅. ԕᔾᰁᰍᄡᙹಽᖁᱶ⦽ᔍŁ

ᝍbࠥ۵Ʊ☖ᔍŁಽᯙ⦽⦝⧕᮹ᱶࠥෝӹ┡ԕ໑, ᫙ᔾᰁᰍᄡᙹ ۵ š⊂ḡ⢽ॅᮡ ᖁᱶ⦹ᩡ݅.

⇵ᱶࡽĥᙹॅ᮹ᔢݡᱢᯙⓍʑෝእƱ⧕ᅙđŝ, ᅙᩑǍᨱᕽ۵

ᵝᵲ༉⩶ᮡࠥಽ᫵ᯙᯕᵝั༉⩶ᨱᕽ۵⪹Ğ᫵ᯙᯕ݅ෙ᫵ᯙᨱ

እ⧕ᔍŁᝍbࠥᨱ޵ฯᮡᩢ⨆ᮥၙ⊹۵äᮝಽӹ┡ԍ݅. ⇵ᱶࡽ

༉⩶ᨱ᮹⦹໕, ᵝᵲ༉⩶ᨱᕽ۵ࠥಽ᫵ᯙᯕᔍŁᝍbࠥᨱၙ⊹۵

ᩢ⨆ᯕaᰆⓑäᮝಽӹ┡ԍᮝ໑, ĥᙹsᯕ0.487ᮝಽ᧲(+)᮹

sᮥaḡ۵äᮝಽӹ┡ԍ݅. ᯕ۵ࠥಽ᫵ᯙᯕ᷾a⧁ᙹಾ, ᷪ

tʙᵝᱶ₉᪡šಉᯩÑӹ, ⠪໕ᖁ⩶ᯕḢᖁ, ࠥಽ⡍ᰆᯕ⎹Ⓧญ✙

ᯝĞᬑᨱᔍŁᝍbࠥa׳ᦥḡ۵äᮥ᮹ၙ⦽݅. ੱ⦽ࠥಽ᫵ᯙ

ᵲ᳦݉Ǎ႑᪡ᔍŁǍeᮡᮁ᮹ᙹᵡ0.05ᨱᕽᮁ᮹⦹ḡᦫᮡäᮝ ಽӹ┡ӹᔍŁᝍbࠥᨱᩢ⨆ᮥၙ⊹ḡᦫ۵äᮝಽӹ┡ԍ݅.

⪹Ğ᫵ᯙᯕᔍŁᝍbࠥᨱၙ⊹۵ᩢ⨆ᮡ0.083ಽԁᦉaาÑӹ

Õ᳑ᯝĞᬑ, ᔍŁᝍbࠥ۵׳ᦥḡ۵äᮝಽӹ┡ԍᮝ໑, ᔍŁ

ၽᔾ᜽e᮹ Ğᬑᨱ۵ ĥᙹ᮹ ᇡ⪙a ᮭ(-)᮹ sᮥ aᲙ ᵝeᯝ

Ğᬑ ᔍŁ ᝍbࠥ۵ ԏᦥḡ۵ äᮝಽ ӹ┡ԍ݅. ᬕᱥᯱ ᫵ᯙ᮹

(6)

Fig. 2. Structural Equation Model for Weekday

Fig. 3. Structural Equation Model for Weekend

ḡ۵äᮝಽӹ┡ԍᮝ໑, ঑௝ᕽᩑಚݡa׳Ł, ݡ⩶₉ෝᬕᱥ⦹۵

ԉᯱᯝĞᬑᔍŁᝍbࠥ۵ԏŁ, ŝᗮᬕᱥᯝĞᬑᨱ۵ᔍŁᝍbࠥ

a ׳ᮡ äᮝಽ ӹ┡ԍ݅.

ၹ໕ᵝั༉⩶ᨱᕽ۵ࠥಽ᫵ᯙᯕᔍŁᝍbࠥᨱaᰆⓑᩢ⨆ᮥ

ၙ⊹۵ᵝᵲ༉⩶ŝݍญᬕᱥᯱ᫵ᯙᯕaᰆⓑᩢ⨆ᮥၙ⊹۵

äᮝಽӹ┡ԍ݅. ੱ⦽⪹Ğ᫵ᯙᯕᬕᱥᯱ᫵ᯙ݅ᮭᮝಽᔍŁ

ᝍbࠥᨱ ⓑ ᩢ⨆ᮥ ၙ⊹۵ äᮝಽ ӹ┡ԍᮝ໑, ᬕᱥᯱ ᫵ᯙ᮹

ĥᙹsᯕ᧲᮹sᮥӹ┡ԕᵝᵲ༉⩶ŝᱶၹݡ᮹᧲ᔢᮥᅕᯕ۵

äᮝಽᇥᕾࡹᨩ݅. ঑௝ᕽᵝั༉⩶ᨱᕽ۵᳦݉Ǎ႑᪡ŝᗮᮡ

᷾a⧁ᙹಾᔍŁᝍbࠥᨱᮭ(-)᮹ᩢ⨆ᮥၙ⊹໑, ԁᦉ, ᵝ᧝, ᬕᱥ

ᯱ ᖒᄥၰᩑಚ, ₉᳦ᮡ᧲(+)᮹ᩢ⨆ᮥၙ⊹۵äᮝಽӹ┡ԍ݅.

ੱ⦽ࠥಽ᫵ᯙᵲ᳦݉Ǎ႑᪡ᔍŁ᭥⊹᮹Ğᬑᵝᵲ༉⩶ᨱᕽ۵

ᔍŁᝍbࠥᨱᮁ᮹⦹ḡᦫᦹᮝӹ, ᵝั༉⩶ᨱᕽ۵ᔍŁᝍbࠥᨱ

ᮁ᮹⦽ ᩢ⨆ᮥ ၙ⊹۵ äᮝಽ ӹ┡ԍ݅.

ԕᔾᰁᰍᄡᙹᯙᔍŁᝍbࠥ۵4}᮹š⊂ᄡᙹᨱ᮹⧕ᖅ໦ࡹ

໑, ᔢݡᱢᯙᩢ⨆ಆᮡᵝᵲ༉⩶ᨱᕽ۵₉ప❭ᗱݡᙹ, ᔍŁ₉పᙹ, ᔍ฾ᯱᙹ, ᇡᔢᯱᙹ᮹ᙽᮝಽ, ᵝั༉⩶ᨱᕽ۵ᇡᔢᯱᙹ, ᔍ฾ᯱ

ᙹ, ᔍŁ₉పᙹ, ₉ప❭ᗱݡᙹ ᙽᮝಽ ӹ┡ԍ݅.

(7)

Table 4. Estimates of 쪢G for Weekday Model

Category Estimates S.E. t-value P Standardized coefficients

Road factors 1.000 - - - 0.487

Environment factors 0.044 0.014 3.225 0.001 0.083

Driver factors -0.163 0.037 -4.407 0.001 -0.082

Table 5. Estimates of 쩊

x

for Weekday Model

Category Observed variables Estimates S.E. t-value P Standardized coefficients

Road factors

Parking on shoulder 1.000 - - - ¿ 0.466

Horizontal alignment 0.783 0.135 5.811 0.001 0.121

Vertical alignment -0.029 0.062 -0.463 0.643 -0.009

Accident location 0.012 0.038 0.307 0.759 0.006

Road surface 0.229 0.130 1.771 0.077 0.035

Environment factors

Weather condition 1.000 - - - 0.595

Daytime or nighttime -0.003 0.025 -0.126 0.900 -0.002

Surface condition 1.051 0.252 4.165 0.001 0.621

Driver factors

Driver’s gender 1.000 - - - 0.257

Driver’s age 3.096 0.209 14.786 0.001 0.218

High operating speed -0.457 0.065 -7.090 0.001 -0.086

Vehicle type 5.936 0.918 6.465 0.001 0.958

Table 6. Estimates of 쪢G for Weekend Model

Category Estimates S.E. t-value P Standardized coefficients

Road factors 1.000 - - - 0.046

Environment factors 0.244 0.067 3.618 0.001 0.115

Driver factors 0.032 0.010 3.232 0.001 0.073

Table 7. Estimates of 쩊

x

for Weekend Model

Category Observed variables Estimates S.E. t-value P Standardized coefficients

Road factors

Parking on shoulder 1.000 - - - 0.067

Horizontal alignment 16.455 7.619 2.160 0.031 0.336

Vertical alignment -4.523 2.163 -2.091 0.037 -0.190

Accident location 3.542 1.661 2.132 0.033 0.232

Road surface 13.802 6.409 2.154 0.031 0.276

Environment factors

Weather condition 1.000 - - - 0.995

Daytime or nighttime 0.033 0.017 1.958 0.050 0.033

Surface condition 1.000 0.018 54.239 0.001 0.996

Driver factors

Driver’s gender 1.000 - - - 0.291

Driver’s age 3.156 0.354 8.920 0.001 0.290

High operating speed -0.197 0.097 -2.032 0.042 -0.047

ᵝᵲ, ᵝั༉⩶ᨱ᮹⧕⇵ᱶࡽ༉ᙹॅၰ⢽ᵡ᪅₉, t-valueෝ

ᱶญ⦹໕ Tables 4~7ŝ z݅.

ᵝᵲᔍŁ༉⩶᮹ Ğᬑ, ⇵ᱶࡽLamda-x᪡Gamma sᨱ᮹⧕,

⠪໕ᖁ⩶, ࠥಽ⡍ᰆ, ࠥಽ⢽໕ᔢ┽, ᬕᱥᯱᩑಚ, ݡ⩶₉పᯝᙹಾ

(8)

Table 8. Goodness of Fit Statistics

Index Weekday Weekend Recommended level

Chi-square 2190.547 (p=0.001) 1034.612 (p=0.001) p<0.05)

GFI 0.968 0.966 0.9 and more

AGFI 0.956 0.953 0.85 and more

RMR 0.012 0.015 0.05 and less

RMSEA 0.051 0.052 0.05 and less

ᔍŁᝍbࠥ᮹᧲(+)᮹ᩢ⨆ᮥၙ⊹໑ŝᗮ॒ᯕᔍŁᝍbࠥᨱᮭ(-) ᮹ᩢ⨆ᮥၙ⊹۵äᮝಽӹ┡ԍ݅. ᵝัᔍŁ༉⩶᮹Ğᬑ⬂݉ᖁ⩶, ᔍŁ᭥⊹, ࠥಽ⡍ᰆᔢ┽, ࠥಽ⢽໕, ᧝e, ݡ⩶₉పᯝᙹಾᔍŁᝍ bࠥ᮹᧲(+)᮹ᩢ⨆ᮥၙ⊹໑, ᳦݉Ǎ႑aᔍŁᝍbࠥᨱᮭ(-)᮹

ᩢ⨆ᮥ ၙ⊹۵ äᮝಽ ӹ┡ԍ݅.

ᅙᩑǍ༉⩶ᯕᵝᨕḥᯱഭᨱᱥၹᱢᮝಽ᯹ᇡ⧊⦹۵ḡෝ❱݉

⦹ʑ᭥⦽ᱢ⧊ࠥḡᙹ۵Table 8ᨱᱽ᜽⦹ᩡ݅. ⋕ᯕᱽŒ☖ĥపᮡ

⢽ᅙŖᇥᔑ⧪಍(S)ŝ⇵ᱶŖᇥᔑ⧪಍(⳸)᪡᮹ᇩᯝ⊹ᱶࠥෝ⠪a

⦹ʑ᭥⦽ḡ⢽ಽ, ᅙ༉⩶ᨱᕽ۵1034.612~2190.547(p=0.001) ಽӹ┡ӹࢱ⧪಍e᮹₉ᯕaᨧ݅۵ȡྕaᖅᮥʑb⦹íࡽ݅.

⦹ḡอ⋕ᯕᱽŒ☖ĥపᮡ⢽ᅙⓍʑᨱၝq⦹íၹ᮲⦹ᩍ⢽ᅙ᮹

ᙹaฯᮥᙹಾpsᯕԏíӹ┡ӹ۵Ğ⨆ᯕᯩᨕ, ݅ෙᩍ్aḡ

ᱢ⧊ḡᙹෝ⧉̹Łಅ⦹ᩍ❱݉⦹۵äᯕၵ௭Ḣ⦹݅. ঑௝ᕽᅙ

ᩑǍᨱᕽ۵ ⋕ᯕᱽŒ ☖ĥప ᯕ᫙ᨱ ᱢ⧊ḡᙹ(goodness-of-fit- index, GFI), ᳑ᱶᱢ⧊ḡᙹ(adjusted goodness-of-fit-index, AGFI),

᯵₉⠪Ɂᯱ᜚᮹ᯕᵲɝ(root mean square residual, RMR), ɝᔍ ᪅₉⠪Ɂᯱ᜚᮹ᯕᵲɝ(root mean square error of approximation, RMSEA)ෝ Łಅ⦹ᩍ ༉⩶᮹ ᱢ⧊ᩍᇡෝ ❱݉⦹ᩡ݅.

ᱢ⧊ḡᙹ(GFI)۵š⊂⧪಍ŝᰍᔾ⧪಍e᮹᯵₉ᯱ᜚⧊᮹እᮉ ᨱʑⅩ⦽ḡᙹಽᕽᯝၹᱢᮝಽ0~1ᔍᯕ᮹sᮥaḡ໑, ༉⩶ᯕ

⢽ᅙŖᇥᔑ⧪಍ᮥᖅ໦⦹۵እᮉᮥӹ┡ԕ۵ḡ⢽ᯕ݅. ᱢ⧊ḡᙹ ᮹ᅕ⠙ᱢᯙǭᰆᙹᵡᮡ0.9 ᯕᔢᯕ໑, ࢱ}᮹ᅙᩑǍ༉⩶༉ࢱ

ʑᵡ⊹ᨱᇡ⧊ࡹ۵äᮝಽӹ┡ԍ݅. ੱ⦽ᯱᮁࠥᨱ᮹⧕ᱢ⧊ḡᙹ

ෝᙹᱶ⦽᳑ᱶᱢ⧊ḡᙹ(AGFI) ᩎ᜽ǭᰆᙹᵡᯙ0.9ᯕᔢᯙäᮝಽ

ӹ┡ԍᮝ໑, S᪡⳸᪡᮹₉ᯕෝᯕᬊ⦹ᩍᔑ⇽⦹۵RMRᮡ༉⩶ᨱ

᮹⦹ᩍᖅ໦⧁ᙹᨧ۵ᇥᔑ/Ŗᇥᔑ᮹Ⓧʑಽᕽ᯲ᮥᙹಾᳬᮡsᮥ

ӹ┡ԕ۵ߑǭᰆᙹᵡᅕ᯲݅ᮡ0.011ŝ0.014ಽӹ┡ԍ݅. ⢽ᅙⓍʑ aⓑ༉⩶᮹⋕ᯕᱽŒ☖ĥపᨱݡ⦽⦽ĥෝɚᅖ⦹ʑ᭥⦽ᱢ⧊ḡᙹ

RMSEA sࠥ bb 0.051 0.052ಽ ᱢ⧊⦹í ӹ┡ԍ݅.

5. đುၰ⨆⬥ᩑǍŝᱽ

ᅙᩑǍᨱᕽ۵ᵝᵲŝᵝัŁᗮࠥಽ᮹ᔍŁᝍbࠥᨱᩢ⨆ᮥ

ၙ⊹۵᫵ᯙॅᯕ₉ᯕaᯩᮭᮥᇥᕾ⦹ŁᯱᔍŁᝍbࠥෝǍ᳑༉

⩶᮹ԕᔾᰁᰍᄡᙹಽ, ࠥಽ᫵ᯙၰᬕᱥᯱ᫵ᯙ, ⪹Ğ᫵ᯙᮥ

᫙ᔾᰁᰍᄡᙹಽᖅᱶ⦹ᩍ༉⩶ᮥ⇵ᱶ⦹ᩡ݅. ᰁᰍᄡᙹॅᮥᖅ໦

⦹ʑ᭥⦽š⊂ᄡᙹᨱ۵tʙᵝᱶ₉, ⠪໕ᖁ⩶, ᳦݉Ǎ႑, ᔍŁǍ e, ࠥಽ⡍ᰆ, ᵝ᧝, ʑᔢᔢ┽, ࠥಽ⢽໕, ᬕᱥᯱӹᯕ, ᬕᱥᯱᖒᄥ, ŝᗮᩍᇡ, ₉᳦, ᔍ฾ᯱᙹ, ᇡᔢᯱᙹ, ᔍŁ₉పᙹ, ₉ప❭ᗱݡᙹ

॒ⅾ16}ᄡᙹa⡍⧉ࡹᨩᮝ໑, 2006֥ᇡ░2010֥ʭḡ᮹Łᗮ

ࠥಽᔍŁߑᯕ░12,042}ෝ⢽ᅙᮝಽᯕᬊ⦹ᩡ݅. ༉⩶᮹⇵ᱶđ ŝ ᵝᵲ ᔍŁ ᝍbࠥ ༉⩶ᮡ ࠥಽ ᫵ᯙᯕ ᔍŁ ᝍbࠥᨱ aᰆ

ⓑᩢ⨆ᮥၙ⊹۵äᮝಽӹ┡ԍᮝ໑, ᵝัᔍŁᝍbࠥ༉⩶ᮡ

ᬕᱥᯱ᫵ᯙᯕᔍŁᝍbࠥᨱaᰆⓑᩢ⨆ᮥၙ⊹۵äᮝಽӹ┡ӹ

ᵝᵲŝ ᵝั ᔍŁ᮹ ᬱᯙ ₉ᯕෝ ӹ┡ԩ݅.

ᵝᵲᔍŁ༉⩶᮹Ğᬑ, ⇵ᱶࡽLamda-x᪡Gamma sᨱ᮹⧕,

⠪໕ᖁ⩶, ࠥಽ⡍ᰆ, ࠥಽ⢽໕ᔢ┽, ᬕᱥᯱᩑಚ, ݡ⩶₉పᯝᙹಾ

ᔍŁᝍbࠥ᮹᧲(+)᮹ᩢ⨆ᮥၙ⊹໑ŝᗮ॒ᯕᔍŁᝍbࠥᨱᮭ(-) ᮹ᩢ⨆ᮥၙ⊹۵äᮝಽӹ┡ԍ݅. ᵝัᔍŁ༉⩶᮹Ğᬑ⬂݉ᖁ⩶, ᔍŁ᭥⊹, ࠥಽ⡍ᰆᔢ┽, ࠥಽ⢽໕, ᧝e, ݡ⩶₉పᯝᙹಾᔍŁᝍ bࠥ᮹᧲(+)᮹ᩢ⨆ᮥၙ⊹໑, ᳦݉Ǎ႑aᔍŁᝍbࠥᨱᮭ(-)᮹

ᩢ⨆ᮥ ၙ⊹۵ äᮝಽ ӹ┡ԍ݅.

Ʊ☖ᔍŁ۵ๅᬑ݅᧲⦽᫵ᯙॅᯕᅖ⧊ᱢᮝಽ᯲ᬊ⦹ᩍၽᔾ⦹

ʑভྙᨱᩩ⊂ᮥ⦹۵ߑฯᮡᨕಅᬡᯕᯩ݅. ✚⯩ӹᅙᩑǍᨱᕽ᮹

ᇥᕾđŝ᪡zᯕᵝᵲŝᵝัᔍŁᝍbࠥᨱᩢ⨆ᮥၙ⊹۵ᵝ᫵⦽

᫵ᯙᮡ݅෕íӹ┡ӽ݅. ᵝᵲᔍŁᨱᕽ۵ࠥಽ᫵ᯙᯕaᰆⓑᩢ⨆

ᯕḡอᵝัᔍŁ۵☖⧪᮹ᵝ᫵༊ᱢᯕᩍaᯕʑ ভྙᨱᬕᱥᯱ

᫵ᯙᯕⓑᩢ⨆ᮝಽᇥᕾࡹᨩ݅. ᵝᵲŝᵝั᮹݅ෙ☖⧪➉▕ᨱ

঑ෙ Ʊ☖ᔍŁ᮹ ᬱᯙᮥ ᇥᕾ⦹Ł ᯕᨱ ঑ෙ ݡ₦ᮥ Łಅ⦹ʑ

᭥⦹ᩍᅙᩑǍᨱᕽᱢᬊ⦽ࠥಽӹᬕᱥᯱ, ⪹Ğ᫵ᯙॅᮥᖙᇥ⪵⦹

Ñӹ, ᝅᱽᱶ₦ᔢฯᯕŁಅࡹḡ༜⦹Łᯩ۵ᬕᱥᯱ᮹⧪┽ᱢ

⊂໕ᮥᰁᰍᄡᙹಽᕽŁಅ⦹ᩍ༉⩶ᮥၽᱥ᜽┉݅໕ᔍŁෝᩩ⊂

⦹Ł ə ⦝⧕ෝ ᵥᯕ۵ߑ ʑᩍ⧁ ᙹ ᯩᮥ äᯕ݅.

qᔍ᮹ɡ

ᯕםྙᮡ2011֥ࠥᱶᇡ(Ʊᮂŝ⦺ʑᚁᇡ)᮹ᰍᬱᮝಽ⦽ǎᩑ

Ǎᰍ݉᮹ ḡᬱᮥ ၼᦥ ᙹ⧪ࡽ ᩑǍᯥ(No. 2011-0028933).

(9)

References

Anowar, S., Yasmin, S. and Tay, R. (2013). “Comparison of crashes during public holidays and regular weekends.” Accident Analysis and Prevention, Vol. 51, pp. 93-97.

principles and practice-, Chungram (in Korean).

Barua, U. and Tay, R. (2010). “Severity of urban transit bus crashes in Bangladesh.” Journal of Advanced Transportation, Vol. 44, No. 1, pp. 34-41.

Gray, R. C., Quddus, M. A. and Evans, A. (2008). “Injury severity analysis of accidents involving young male drivers in Great Britain.” Journal of Safety Research, Vol. 39, pp. 483-495.

Lee, J. Y. (2008). “Analysis of traffic a accident size for Korean highway using structural equation models.” Accident Analysis

and Prevention, Vol. 40, pp. 1955-1963.

Lee, J. Y., Chung, J. H. and Son, B. S. (2008). “Analysis of traffic accident severity for korean highway using structural equations models.” Journal of Korean Society of Transportation, Vol. 26, No. 3, pp. 17-24 (in Korean).

Park, H. S., Son, B. S. and Kim, H. J. (2007). “Development of accident prediction models for freeway interchange ramps.”

Journal of Korean Society of Transportation, Vol. 25, No. 3, pp.

123-136 (in Korean).

Park, J. S., Kim, T. Y. and Yu, D. S. (2007). “Correlation analysis and estimation modeling between road environmental factors and traffic accidents (the case of a 4-legged signalized intersections in cheongju).” Journal of Korean Society of Transportation, Vol.

25, No. 2, pp. 63-72 (in Korean).

(10)

수치

Fig. 1. Structural Equation Modelḡᩎᱢ ⬉ŝ, ĥᱩᱢ ⬉ŝ, ᵝᵲ/ᵝั, ʑᔢᔢ┽, ࠥಽ ⢽໕ᔢ┽
Table 2. Descriptive Statistics of Accident Records
Table 3. Definition of Model Variables
Fig. 2. Structural Equation Model for Weekday
+3

참조

관련 문서

Through the statistical analysis of marine accident reports, human factors which contribute to accidents is defined as the classification of accident causes

The optimized STL files of the 3D tooth models were imported into the finite element analysis software, CosmosWorks (Structural Research &amp; Analysis

period was prolonged unavoidably, (3) by explaining the risk factors associated with the failure to patients honestly, and subsequently performing

The major contributions of this study are identifying the influence on the performance of venture business by means of structural equation model and the

그림에서 보인 들보를 Gauss-Siedal Method 기초한 모멘트 분배법에 의하여 해석 하고 B 점과 C 점에서의 회전각을

The Medical accident and the

Global structural analysis model of the 300K VLCC by using the developed method. (Shell element, 1-D

1) Derive shear flow dispersion equation using Taylor’ analysis (1953, 1954) - laminar flow in pipe (1953).. -