Sirmans and Swicegood (1997, 2000) also test experience measured by years with thecurrent firm (Expf), and therefore, this measure of experience is included here as well.3. Similar to Follain, Lutes and Meier (1987), gender is not statistically significant in themodel; therefore, it is not included in the regression.4. These variables correspond with the following questions shown in the survey instrument(Appendix 1):WebQ32,WebpQ43,WeblistQ33,InternetQ40,EmailQ39,TechQ35,SoftQ37 andCageQ36.5. An itemized listing of the survey responses to each question on the survey questionnaireis available from the authors.6. The second factor has an eigenvalue of 1.46 and an explained variation of 18.21%, whilethethirdfactor hasaneigenvalueof1.02andanexplainedvariation of 12.80%.Unfortunately, the pattern relating the eight technology variables has no logical explanation,and neither of these factors was statistically significant in the regression analysis.7. The mean of the primary information is.018 rather than 0.0 because the factor analysiswas conducted using 185 observations (all of the survey respondents that answered theinformation technology questions). The regression analysis was conducted using only 149survey respondents, because of other missing or incomplete data. When calculated over thesmaller sample, the mean of the information technology factor is0.018.8. This income elasticity coefficient can vary considerably; Sirmans and Swicegood report thiscoefficient to be 0.910 in their 1997 study. As a comparison, Crellin, Frew and Jud (1988)report this coefficient as 0.634 for a national sample.9. The possibility that the broker variable was capturing variation attributable to being themajority owner of a firm (see question #28 in Appendix 1) was also tested. However, avariable for majority ownership was not statistically significant in the regression equation.The estimated percentage change for theBrokvariable as reported in Exhibit 4 can bedetermined by the following transformation: exp(D)1, whereDis the dummy variablecoefficient.10. When the relationship between technology use and the other independent variables in themodel was examined, the use of information technology (Techf) was found to besignificantly related to hours-worked, schooling and franchise affiliation. Interestingly, therewas no significant association with age, gender, race, buyer-brokerage or possession of abrokerage license.11. A similar analysis using the mean and standard deviation could be conducted for the otherindependent variables; this approach is reasonable when the units of measurement do notinherently offer an intuitive explanation. Other variables such as years of schooling andexperience do not have the same interpretation problems.ReferencesBarta, P., Realtors Invest in Job Security on the Internet,The Wall Street Journal, (Monday)November 6, 2000, B1.
16JOURNAL OF REAL ESTATE PRACTICE AND EDUCATIONVOLUME 5, NUMBER 1, 2002Becker, G. S.,Human Capital, Second edition, New York, NY: Columbia University Press,1975.
You've reached the end of your free preview.
Want to read all 17 pages?
- Fall '13