Quantitative research into the effects of new attractions: error correction model
On the basis of a study of literature we have discovered the variables which could influence the number of visitors to theme parks. We started of with the general tourist literature and subsequently researched specifically what knowledge was available there regarding attraction and theme parks. The list of relevant variables has been worked out further through talks with five experts of the field. Based on this research we found the following variables to be important: income (Crouch, 1994; Witt & Witt, 1995; Lim, 1997; SCP, 2004), price (Schwagermann, 1991; TR&M, 1993; Thach & Axinn, 1994; Kemperman, Borgers, Oppewal & Timmermans, 2000; Swarbrooke, 2002; Braun & Soskin, 2003); cost of travelling (Braun & Soskin, 2003; Wanhill, 2003); marketing[1] (Richards, 1992; Davidson, 1998; TR&M, 1998; Swarbrooke, 2002), weather (McClung, 1991; Schwagerman, 1991; SEO, 1995; ING, 2002; NRIT, 2005) ; weekend days, national holidays and vacations (Schwagerman, 1991; SEO, 1995; Krider & Weinberg, 1998; Corning & Levy, 2002; ING, 2002; NRIT, 2005)
The park of which the data have been processed is situated in the Northern part of Europe and has a visitors number of more than one million (but less than four million) every year. Data have been gathered from the period between the 1st of April 1998 until the 30th of October 2007. In this period the park opened 8 new attractions. Data have been processed on a daily basis. Which means that from the 1st of April we have determined the number of visitors on that specific day, what the average temperature was on that day, whether it rained that day and how much, what was the entrance fee[2] etc.
In table 1-3 you will find the output of the definite model. The adjusted R Square of the model is 64,6%. This means that almost two third of the variance of difference on daily visitor numbers can be explained by the variables in the model. As table 3 shows the average long-term attraction-effect of the investigated park is 10%. This can be calculated by dividing the B-score of NEWATTR_t_1 by the adjustment parameter (-,422) and take the antilog from that result. As can be read from this table the impact of this variable is significant at a p<.10 level. The result can be interpreted as multiplier. This means that introducing a new attraction in this park has an average impact of 10% more visitors, compared to ……
