This occurs when the appraiser or the valuation (i.e. the techniques, processes, systems, etc. used by the appraiser) exhibit random or systematic errors. A review of the literature indicates that no study has investigated the influence of a decision support tool in property valuation in the context of the South African property market. Of utmost importance is the use of computer technology to automate much of the decision analysis.
These attributes determine the possible effect of the DSS on users' decision-making processes: what users can and will do with the system. The lack of understanding of the psychological effects and the integration of contemporary research on behavioral decision making limits the application and usefulness of decision support tools (Elam et al., 1992). Decision Support Systems and the Cost-Benefit Framework of Cognition The theory states that decision making depends on the effort required to make a decision and on the accuracy of the outcome (Payne, 1982).
However, the use of a computer-based DSS did not support the assumption that using an automated system would reduce the strength of the anchoring and adjustment bias. In this version, the program does not support decisions that occur at different levels of the decision-making process. However, it was considered an indicator that the anchoring effect was present when a participant adjusted the value as soon as the alert function was displayed in the DSS version of the experiment.
A random sampling method was used to distribute one of the two versions of the valuation tool to the treatment and control groups.
Analysis of Data and Discussion 1. Data collection
To increase response rates, study participants could win an iPad or one of three iPods through a random drawing. Either the modified robust Levene-type test or the modified robust Brown-Forsythe Levene-type test from the mean with the modified zero-correction method can be applied. However, for unequal and small sample sizes, the modified robust Brown-Forsythe Levene type test is preferred.
For the assumption of homogeneity of variance not to be violated, a significance level greater than 0.05 must be reached (H0: VarSTD = VarDSS, p-value > 0.05). Using three measures of variation, the null hypothesis must be rejected if the majority of the measures show a higher variation for DSS than for STD. Similarly, the standard deviation is slightly higher for the STD version (=2.16) than for the DSS version (=1.98) and is confirmed by the coefficient of variation.
To test the significance level of the results, the jb test was used to examine the normality of the data. Interestingly, the DSS version was normally distributed, while the STD data and the overall sample were far from a Gaussian distribution. Given the unbalanced and small sample sizes of the data, the modified robust Brown-Forsythe Levene type test of the mean with the modified null correction method was applied (Table 4).
The same measures of variation and significance level tests were used to test the effectiveness of the DSS version within the subsamples of experts and students. As shown in Table 5, measures of variation for both groups were higher under the STD version than under the DSS version. The coefficient of variation indicates that the dispersion under the STD version was higher than under the DSS version.
The coefficient of variation was also larger in the STD version, indicating a higher spread of results than for the DSS version. The DSS version produced fewer values of estimation variation and outliers in the expert group were less extreme than with the standard versions. In a similar way, it can be concluded that some members of the test groups were sensitive to the effect of anchoring and adjustment.
Conclusions
The second hypothesis states that the anchoring effect is reduced when the evaluator is unbiased and supported by his decision. Similar to Lausberg and Dust's (2017) observations, the illogical adaptation may be due to the fact that the warning message was not clear enough and may have confused the participants. After adjusting the valuation results with positive or no adjustment, it was found that only seven participants (=22%) positively adjusted the market value.
When comparing the tools, descriptive statistics showed that the spread was more frequent in the STD version than in the DSS version. Statistical tests showed no significance at the 0.5 level that the valuation variances would be reduced with the given DSS tool. A similar observation, at the 0.05 significance level, was made during Lausberg and Dust's (2017) experimental research.
However, unlike the German study that showed some evidence of benefits of the DSS tool at the 1% level of significance, this study could not support similar results. This can be explained by the fact that, unlike the previous study, which showed that German evaluators were not aware of decision-making during evaluation tasks (Lausberg and Dust, 2017), in the present situation, African test subjects Souths were perhaps more conscious of providing value. judgments. Basic descriptive statistical measures show some evidence that decision support tools can help avoid decisions.
Although the significance test did not fully support the effectiveness of the DSS instrument, it is noted that a decision support system can provide better results than the standard instrument at different decision levels of the valuation process. There was also evidence for the anchoring-adaptation heuristic, and it was observed that the automated system can help counteract the unwanted cognitive mechanism generated by inexperienced decision makers.
Implications for Valuation Practice and Research
Third, the experiment should be repeated with experts of different levels of experience and expertise, for other types of properties and using different valuation methods. Other forms of heuristics, such as representativeness and availability heuristics, should be included in the experiment to determine their impact on evaluation results. Furthermore, the experiment should be extended to other debian methods, especially process changes and training.
In our opinion, this type of software can be a useful supplement to existing procedures, it is not intended to replace the appraiser. Fourth, the developers of valuation software must get involved in the further development of their products in the direction of decision support. So far, most programs are better calculators that do not support the appraiser in his decisions.
However, we believe that decision support is both a key to improving the valuation quality and a way for traditional valuation tools to differentiate themselves from the automated valuation models, which are superior to them in efficiency but often lack efficiency. Behavioral disputes that have been addressed for many years in the real estate literature must finally be discussed and presented to students as well as to experts in general. Amidu (2011) highlights the needs for property valuation education, improvement in professional standards, a code of conduct and accountability to help counter and possibly overcome dysfunctional behavior in value judgment tasks.
Acknowledgments
How appraisers do their job: a test of the appraisal process and the development of a descriptive model. When effortful thinking affects judgmental anchoring: Differential effects of Forewarning and Prompts on self-generated and externally provided anchors. An experimental evaluation of the effect of data presentation on heuristic bias in commercial valuation.
Paper presented at the 12th Pacific Rim Real Estate Society Conference, Auckland, New Zealand, 22-25 January. An examination of valuation bias: The role of decision support tools in deductive valuation judgments. An experimental study of the impact of computer-based decision aids on decision-making strategies.
Appendix A – Assessment Tools Home Page, for both Standard Programs (STD) and Decision Support Programs (DSS).
Appendix A – First page of the Valuation Tools, for both Standard (STD) and Decision Support (DSS) Programmes
Appendix B – Third page of the valuation tools, for both Standard (STD) and Decision Support (DSS) programs.
Appendix B – Third page of the valuation tools, for both Standard (STD) and Decision Support (DSS) Programmes
Appendix C – Second page for Standard (STD) Programme Only
136 Appendix D - Page Two for Decision Support Program (DSS) Only .. Market Value Calculation as of August Data and Assumptions. First, estimate the total cost using the cost-to-income ratio. Now divide the total cost into different costs by overwriting the percentages in the dark green boxes.
If you think the current portions are OK, you can leave them as they are. For the next step, you can look at the market data in the text. In this section, the program supports your data entry and calculation of market rents and other factors.
In addition, you can now rate the property against the market on a 5-point scale. The chart shows the range of asking prices for 9 nearby properties as a vertical black line and the average as a horizontal green line. When you have filled in all the information and you are satisfied with the result, click on "Continue".
