본 연구는 호텔웨딩연회의 혼잡지각에 따른 감정반응이 행동의도에 미치는 영향을 실증분석 하였고, 조사 대상은 웨딩연회를 이용하고 경험한 고객을 대상으로 총 450명을 대상으로 설문을 실시하였으며 기간은 2013년 3월에서 5월까지 약 12주간에 걸쳐 진행되었다. 설문지는 450부를 배포하여 426부를 회수하였으며 내용이 불성실하거나 통계적으로 활용이 불가능하다고 판단되는 24부를 제외하고 최종 402부를 실증분석에 활용하였다. 회수된 설문지는 SPSS와 AMOS 통계패키지 프로그램을 활용하여 인구통계특성은 빈도분석을, 가설검증을 위한 기초분석은 구성개념의 내적 일관성과 신뢰성 검증을 위해 사용되고 있는 크론바흐 알파계수분석(Chronbach''s Alpha Analysis)과 변수의 구조를 파악하고 타당성검증을 위해서 요인분석(Exploratory Factor Analysis)과 확인적 요인분석(Confirmatory Factor Analysis), 관련성의 유무를 위해서는 상관관계분석(Correlate Analysis), 모형간의 인과관계를 파악하기 위해 구조방정식모형(Structural Equation Model)등을 적용하여 실시하였다.
가설 검증결과 총 8개의 가설 중 7개의 가설이 채택되어 혼잡지각이 감정반응과 행동의도에 미치는 영향과의 관련성이 검증되었다.
연구가설 검증결과를 요약하면 다음과 같다.
첫째, 변수들의 내적 일관성과 신뢰성을 분석하기 위하여 Cronbach''s α 계수를 이용하여 분석한 결과 인적혼잡성, 공간적 혼잡성, 긍정적 감정반응, 부정적 감정반응, 고객만족, 추천의도에 각각 .881, .849, .859, .867, .779, .688로 비교적 높게 나타나 각 변수들의 신뢰성은 확인되었다.
둘째, 인적 혼잡성은 긍정적 반응과 부정적 반응에 영향을 미치는 영향에 대한 가설 분석결과 영향력은 .166, -.117로 나타났으며 t 값은 2.823, -2.062로 나타나 기각비의 절대 값이 ±1.96이상 이므로 유의확률 p <0.05 에서 유의하게 나타나 가설 1과 가설 2는 채택이 되었다.
셋째, 공간적 혼잡성은 긍정적 반응과 부정적 반응에 -.352 .487로 나타났으며 t 값은 -6.569, 8.863으로 나타났으며 유의수준 .000으로 나타나 가설 3과 가설 4는 채택이 되었다.
넷째, 긍정적 반응은 고객만족과 추천의도에 미치는 영향에 대한 분석 결과 영향력은 .431 과 -.080으로 나타났고 유의수준 t 값은 7.744와 -1.500으로 긍정적 반응은 고객만족에는 유의하나 추천의도에는 유의한 영향이 미치지 않고 기각으로 나타나 가설 5는 채택이 되었고 가설 6은 기각이 되었다.
다섯째, 부정적 반응은 고객만족과 추천의도에 각각 t 값이 -3.288과 -6.079로 나타나 유의수준 .001과 .000에서 유의하게 설명되고 있어 가설 7과 가설 8은 채택이 되었다. 끝으로 고객만족에는 유의한 영향을 미치나 추천의도에는 기각된 결과를 나타내었다. 이는 호텔웨딩연회 특성상 장소나 주최자의 선택이 다르게 적용되므로 추천효과로 인한 선택은 적용되지 않고 있음을 알 수 있다.
본 연구의 한계점으로 혼잡지각에 대한 측정도구를 개발하기 위해 선행연구를 바탕으로 혼잡지각에 대한 설문항목을 호텔웨딩연회 특징에 맞도록 설계하도록 시도하였으나, 고객의 일반특성에 따른 연구, 즉 연령별에 따른 기혼자와 미혼자 또는 잠재고객과 기존고객과의 관계, 연회공간의 특수성과 연회형태에 대한 관계를 설명하기에는 아쉬움이 있다.

This study made an empirical analysis on the effects of emotional responses resulted from crowding perception of hotel wedding convention on behavioral intentions. And subjects surveyed were customers using and experiencing wedding convention and a survey was conducted with 450 people in total. It was carried out for around 12 weeks from March to May in 2013. 450 questionnaires were distributed and 426 were collected. Finally 402 questionnaires were employed for our empirical analysis excluding 24 which were decided to be untrustworthy in their content or to be impossible to use statistically. For collected questionnaires, frequency analysis was made to investigate demographic characteristics using SPSS and AMOS statistical package program. For elementary analysis for hypothesis testing, Chronbach''s alpha analysis was made, which is used for testing reliability and internal consistency of a construct, exploratory factor analysis and confirmatory factor analysis was done for testing validity, correlate analysis was done for testing the status of relationships, and structural equation Model was applied to understand casual relationships between models.
As a result of hypothesis testing, 7 out of 8 hypotheses in total were adopted to test relationships with the effects of crowding perception on emotional responses and behavioral intentions.
Results of testing research hypotheses are summarized as follows:
First, as a result of analyzing using Cronbach''s α coefficient in order to analyze internal consistency and reliability of variables, the reliability of each variable were confirmed as human crowding, spatial crowding, positive emotional response, negative emotional response, customer satisfaction, and intention to recommend were .881, .849, .859, .867, .779, and .688 respectively, which were relatively high.
Second, as a result of analyzing a hypothesis of the effect of human crowding on positive and negative responses, it was found that the effect was .166, and -.117 respectively, and t value was 2.823, and -2.062 respectively. As the absolute value of dismissal ratio is more than ±1.96, it has significance in significance probability p <0.05 and hypotheses 1 and 2 were adopted.
Third, spatial crowding was -.352 and .487 respectively in positive response and negative response and t value was -6.569 and 8.863 respectively and the significance probability was .000 and .000 and hypotheses 3 and 4 were adopted.
Fourth, for positive response, as a result of analyzing its effect on customer satisfaction and intention to recommend, the effect was .431 and -.080 and significance level t value was 7.744 and -1.500, which indicates that positive response has no significant effect on intention to recommend while it has significance in customer satisfaction being dismissal and hypothesis 5 was adopted and hypothesis 6 was rejected.
Fifth, for negative response, t value was -3.288 and -6.079 respectively in customer satisfaction and intention to recommend and significance level .001 and .000 was explained significantly and hypotheses 7 and 8 were adopted. Finally, it has a significant effect on customer satisfaction but showed a result of refusal in intention to recommend, which indicates that choice resulted from recommendation effects is not applied since choice of organizers or places are applied differently because of the nature of hotel wedding convention.
Concerning limitations of the current study, we attempted to design items of questions about crowding perception, which match with the nature of hotel wedding convention based on precedent studies in order to develop measuring scales of crowding perception, but it can be said that they are unable to explain the relationships with the married and singles according to age, which means a study according to general characteristics of customers, or the relationships with potential customers and the existing customers.