INTRODUCTION

Accounting practitioners and researchers recognize the need for reliable audit tools to assist auditors in their evaluation of the going concern question. The going concern assumption is fundamental to the preparation of financial statements in accordance with generally accepted accounting principles. The postulate states that, in the absence of evidence to the contrary, the firm should be viewed as remaining in operation indefinitely (AICPA 1988b). Generally, the auditor does not encounter any unusual audit opinion problems in situations where the going concern assumption is valid. However, when the continued existence of a firm is in question, the auditor is faced with potentially difficult decisions related to the audit opinion. SAS No. 59 is the current authoritative guidance available to help auditors assess the going concern issue.

SAS No. 59 requires auditors to take an active role in their evaluation of a company’s ability to continue as a going concern. Raghunandan and Rama (1995) suggest that the increased responsibilities of auditors also increased the costs associated with issuing NGCOs to companies that subsequently fail. For example, auditors may have greater difficulty defending against lawsuits by investors and creditors when companies fail after receiving NGCOs from auditors.

Prior to SAS No. 59, the authoritative guidance to help auditors evaluate going concern issues was SAS No. 34 entitled, "The Auditor’s Consideration When a Question Arises About an Entity’s Continued Existence." Under SAS No. 34, the auditor had a passive responsibility in assessing an entity’s continued existence. That is, the auditor was required to assess the firm’s going concern status only when contrary information was discovered during the audit of the financial statements. If, after assessing a company’s going concern status, the auditor had both substantial doubt and questions about the recovery of recorded asset values, the auditor was required to modify the audit opinion. No modification was required if the auditor had only substantial doubt about the company’s ability to continue as a going concern.

Though SAS No. 59 increased the auditor’s responsibilities, it did not specify audit procedures that auditors could use to evaluate the going concern assumption. However, the standard used analytical procedures as an example of audit procedures that may identify conditions indicative of possible substantial doubt on the part of auditors about a company’s ability to continue as a going concern. Additionally, in April 1988 the ASB issued SAS No. 56 entitled, "Analytical Procedures," which formally required auditors to use analytical procedures in all financial audits. SAS No. 56 did not set forth analytical procedures that auditors should use in their evaluation of the going concern issue; however, bankruptcy prediction models have been linked to this evaluation (Hopwood, McKeown, and Mutchler 1994; Blocher and Loebbecke 1993; Altman 1993; Koh 1991; Mckee 1989; and Dugan and Zavgren 1988).

The objective of this study is to assess the usefulness of Zmijewski’s, Ohlson’s, and Altman’s bankruptcy prediction models in identifying companies with financial conditions that warrant GCOs after SAS No. 59. The prediction models may alert auditors to certain problems that are difficult to detect with traditional auditing procedures. If the Zmijewski, Ohlson, and Altman models are useful audit tools for evaluating a firm’s going concern potential, then the models should be considered by auditors in making GCO decisions. However, the potential usefulness of the models may have declined subsequent to SAS No. 59 relative to other procedures available to auditors for detecting going concern problems. The Zmijewski (1984), Ohlson (1980), and Altman (1968) models are referred to as the X, Y, and Z-score models throughout this paper.

The Cohen Commission and pre-SAS No. 59 studies [i.e., studies that used data from periods prior to January 1, 1989, the effective date of the standard] suggest that bankruptcy model predictions are more accurate than auditor opinions in signaling impending failure (Koh 1991; Altman 1982; and Altman and McGough 1974). The auditors’ (models’) accuracies for signaling impending failure ranged from 40% to 54% (82% to 93%) in pre-SAS No. 59 studies. The ASB increased the auditors’ responsibilities for the going concern evaluation in SAS No. 59. Accordingly, auditors may have become more accurate at signaling impending failure after SAS No. 59 was issued. This study evaluates whether the gap between the accuracy rates of auditors and those of bankruptcy prediction models narrowed subsequent to SAS No. 59. Evidence that the gap narrowed would suggest that the models continue to be a useful audit tool for auditors when they are evaluating the going concern question.

The Link between Bankruptcy Prediction Models and the Going Concern Evaluation

The auditor’s assessment of the going concern issue is a complex process that can benefit from the use of a decision aid (Paquette and Skender 1996). Altman and McGough (1974) suggested that bankruptcy prediction models may help auditors judge companies’ abilities to continue as a going concerns by alerting auditors to certain problems that may be difficult to detect using traditional auditing procedures. The Cohen Commission also indicated that statistical failure models might very well be considered by auditors in their overall assessments of companies (Commission 1978). Other evidence that bankruptcy prediction models may be useful to auditors in making going concern judgments was provided by Hopwood et al. (1994), Koh (1991), Levitan and Knoblett (1985), Altman (1982), and Deakin (1977). Additionally, the Proceedings of the Expectations Gap Roundtable called for continued research on the effectiveness of analytical procedures in various contexts, including the going concern evaluation (Blocher and Loebbecke 1993). These proceedings specifically identified the use of bankruptcy prediction models as a potential analytical procedure for evaluating the going concern question.

Though there is support for the use of prediction models in the going concern evaluation, the Zmijewski (1984), Ohlson (1980), and Altman (1968) models have not been evaluated in this context subsequent to SAS No. 59.

Altman and McGough (1974) provided a link between bankruptcy prediction models and auditors’ opinion decisions by comparing the accuracy of Altman’s (1968) bankruptcy prediction model to auditors’ opinions prior to the bankruptcy event. They analyzed the model’s predictions and auditors’ opinions for 34 firms that filed bankruptcy during the 1970-1973 period. The results indicated that the Z-score model correctly signaled impending failure prior to bankruptcy in 82% of the cases. They reported that auditors’ opinions signaled impending failure in only 46% of the cases.

Altman (1982) extended the evaluation of Altman’s (1968) model in the auditors’ opinion context using two additional samples: (1) 37 bankrupt firms from 1974-1978 and (2) 44 bankrupt firms from 1978-1982. The Z-score model correctly signaled impending failure for 81.1% (93%) of the 1974-1978 (1978-1982) companies; additionally, he reported that auditors issued GCOs to 59.5% (40%) of the 1974-1978 (1978-1982) companies. Combining the results of the Altman and McGough (1974) and Altman (1982) studies, the Z-score model (auditors) provided early warning signals of subsequent failure in 86.2% (48.1%) of the cases.

These results are somewhat dated since the samples used to evaluate the model were selected from the pre-SAS No. 59 period. Furthermore, prior studies have not evaluated the use of bankruptcy prediction models as audit tools for assessing the going concern issue using data subsequent to the issuance of SAS No. 59. The ASB argued that, at one extreme, all research performed prior to the issuance of SAS No. 59 is of only historical interest since the standard significantly changed (Carmichael and Pany 1993). Even so, both studies concluded that Altman’s (1968) model was a useful tool for auditors’ going concern evaluations. Additionally, the results supported the notion that bankruptcy prediction models are better than auditors at signaling the future prospects of companies.

Carmichael and Pany (1993) indicated that auditors’ failures to issue GCOs to bankrupt companies were at the heart of the expectations gap between auditors and financial statement users. SAS No. 59 charged the auditor with an affirmative responsibility for investigating the going concern status of a firm; consequently, it is questionable whether auditors’ opinions continue to be inferior to bankruptcy prediction models at providing early warning signals of impending bankruptcies after the more stringent standard was issued. This study evaluates whether the gap between the models’ and auditors’ accuracies for signaling impending failure narrowed subsequent to SAS No. 59. 

Specifically, this study evaluates whether auditors’ GCO decisions are more consistent with the models’ predictions for bankrupt companies after the issuance of SAS No. 59. Finding that the gap has narrowed would suggest that the ASB’s efforts to increase auditors’ responsibilities in the going concern evaluation have been effective.

Prior studies used restricted samples, that included only bankrupt companies, to evaluate bankruptcy prediction models in the auditors’ opinion context (e.g., Hopwood et al. 1994, Koh 1991, Levitan and Knoblett 1985, Altman 1982, Deakin 1977, and Altman and McGough 1974). Arguably, firms that declare bankruptcy should have received GCOs; however, firms that receive GCOs do not always file for bankruptcy. Auditors are confronted with decisions of whether to issue GCOs to firms from a variety of financial distress situations, not just bankruptcies. This study evaluates the ability of the Zmijewski, Ohlson, and Altman bankruptcy prediction models to foreshadow GCOs for financially distressed companies other than bankruptcies.

  Specifically, this study evaluates whether auditors’ GCO decisions are more consistent with the models’ predictions for financially distressed companies, other than bankruptcies, after the issuance of SAS No. 59.

This section explains the methodology employed to evaluate the usefulness of the X, Y, and Z-score models as audit tools in going concern assessments. Specifically, the models and the sample are discussed as well as the tests used to evaluate the hypotheses.

Zmijewski (1984) used financial ratios that measured firm performance, leverage, and liquidity to develop his model. The ratios were not selected on a theoretical basis, but rather, on the basis of their performance in prior studies. Zmijewski estimated the model using probit analysis, which weights the log-likelihood function by the ratio of the population frequency rate to the sample frequency rate of the individual groups, bankrupt and nonbankrupt. Zmijewski’s probit model based on 40 bankrupt and 800 nonbankrupt industrial firms was

X = - 4.3 - 4.5 X₁ + 5.7 X₂ - .004 X₃ (1)

where:

X₁ = net income/total assets;

X₂ = total debt/total assets;

X₃ = current assets/current liabilities;

X = overall index.

Zmijewski (1984) developed numerous models using 40 bankrupt and 40 to 800 nonbankrupt firms; however, the model based on the 40:800 proportion of bankrupt to nonbankrupt firms is the model most frequently used by accounting researchers (e.g. Carcello et al. 1995 and Chen and Wei 1993).

Ohlson (1980) selected the predictors he used to develop his model because they appeared to be the ones most frequently mentioned in the literature. He used logistic analysis to derive his bankruptcy prediction model using nine measures of firms’ size, leverage, liquidity, and performance. Based on a sample that included 105 bankrupt and 2,058 nonbankrupt industrial firms, his model was

Y = -1.3 - .4 Y₁ + 6.0 Y₂ - 1.4 Y₃ + .1 Y₄ -2.4 Y₅ - 1.8 Y₆ + .3Y₇ -1.7 Y₈ - .5Y₉ (2)

where:

Y₁ = log(total assets/GNP price-level index);

Y₂ = total liabilities/total assets;

Y₃ = working capital/ total assets;

Y₄ = current liabilities/current assets;

Y₅ = one if total liabilities exceed total assets, zero otherwise;

Y₆ = net income/ total assets;

Y₇ = funds provided by operations/total liabilities;

Y₈ = one if net income was negative for the last two years, zero otherwise;

Y₉ = measure of change in net income;

Y = overall index.

To develop the Z-score model, Altman (1968) compiled a list of twenty-two financial ratios and classified each into one of five categories (liquidity, profitability, leverage, solvency, and activity). Altman selected the ratios on the basis of their popularity in the literature and his belief about their potential relevancy to bankruptcy. He estimated the model using multiple discriminant analysis, which attempts to derive a linear combination of variables that best discriminates between bankrupt and non-bankrupt groups. After numerous tests, the linear function that best discriminated between the 33 bankrupt and 33 non-bankrupt manufacturing firms was:

Z = 1.2 Z₁ + 1.4 Z₂ + 3.3 Z₃ + .6 Z₄ + .999 Z₅ (3)

where:

Z₁ = working capital/total assets;

Z₂ = retained earnings/total assets;

Z₃ = earnings before interest and taxes/total assets;

Z₄ = market value equity/book value of total debt;

Z₅ = sales/total assets;

Z = overall index.

This Z-score model is still cited and used by accounting researchers, practitioners, and educators more than any other bankruptcy prediction model (Altman 1993).

Though the Zmijewski, Ohlson, and Altman models are the only models evaluated in this study, the findings of this study may apply to other models that were derived using a similar methodological process.

The analyses in this study used a 1985-1987 sample and a 1988-1991 sample, with each sample including distressed firms. Distressed companies were from Compustat and at least fulfillment of one of the below is required:

• Chapter 11 Bankruptcy: Compustat’s Industrial Annual Research File (CIAR) contains companies that were deleted from the Industrial Annual File (CIA) because of bankruptcy. CIAR identifies bankrupt firms with an 02 code for footnote 35.
• Chapter 7 Liquidation: CIAR also contains companies that were deleted from CIA because of liquidation. CIAR identifies liquidated firms with an 03 code for footnote 35.
• Bonds vulnerable to default: Both CIAR and CIA report bond ratings for companies evaluated by S&P. Companies with bonds rated CCC or below were included in the sample. These companies were identified by codes 19-24 for data item 280.
• Low stock rating: CIAR and CIA also report stock ratings for companies evaluated by S&P. Companies whose stock was rated as "lower B" and below were included in the sample. These companies were identified by codes 18-22 for data item 282.

The final 1985-1987 (1988-1991) sample included 153 (161) distressed companies. The financial ratios described in equations (1), (2), and (3) were calculated for each firm in the samples using data from CIAR and CIA. Analyses in this study distinguish between bankruptcy and other financially distressed companies. Thus, the 1985-1987 and 1988-1991 samples were partitioned into two categories: (1) those identified as distressed because of bankruptcy; (2) those identified as distressed for reasons other than bankruptcy. Table 1 reports that the 1985-1987 (1988-1991) samples included 103 (108) bankrupt companies and 50 (53) companies that were identified as financially distressed because of reasons other than bankruptcy.

The companies’ audit opinions were identified using codes for auditors’ opinions reported on CIAR and CIA. Companies with GCOs were defined as those reported by Compustat as meeting one of the following conditions:

• Unqualified opinion with explanatory language: Both CIAR and CIA code data item 149 as 4 when auditors expressed an unqualified opinion regarding the financial statements by adding explanatory language to the standard report.
• Disclaimer: Both CIAR and CIA code data item 149 as 3 when auditors refused to express opinions regarding companies’ abilities to sustain operations as going concerns.

Table 2 reports the distribution of the GCOs by type. SAS No. 59 indicates that when auditors have substantial doubt about companies’ abilities to continue as going concerns, they are required to issue either unqualified with an explanatory paragraph or disclaimer opinions. It should be noted that virtually no authoritative guidance or published research exists that auditors could use to determine which type of opinion to issue (LaSalle and Anandarajan 1996). Additionally, LaSalle and Anandarajan (1996) indicated that no evidence exists to suggest that the differences in auditors’ reporting decisions related to going concern decisions are systematic. Thus, this study included both unqualified with an explanatory paragraph and disclaimer opinions as GCOs. Table 2 shows that substantially all of the GCOs used in this study were unqualified opinions with explanatory paragraphs.

The remainder of this section discusses the tests used in this study. Specifically, the tests used to evaluate the X, Y, and Z-score models as audit tools in the going concern judgment are discussed.

This study evaluated the correlation between the X, Y, and Z-score models’ predictions and auditors’ opinions before and after the issuance of SAS No. 59. The correlation between the models’ predictions and auditors’ opinions was evaluated using the following sample subsets. (1) pre-SAS No. 59 bankruptcies (pre-B): Subset of the 1985-1987 sample containing only bankrupt companies; (2) post-SAS No. 59 bankruptcies (post-B): Subset of the 1988-1991 sample containing only bankrupt companies; (3) pre-SAS No. 59 financial distress (pre-FD): Subset of the 1985-1987 sample containing only financially distressed companies other than bankruptcies; and, (4) post-SAS No. 59 financial distress (post-FD): Subset of the 1988-1991 sample containing only financially distressed companies other than bankruptcies. It should be noted that nondistressed companies were not included in these analyses since prior studies have shown that models and auditors rarely issued GCOs to healthy firms. For example, both Levitan and Knoblett (1985) and Koh (1991) reported that the models and auditors correctly classified 100% of the nonbankrupt firms as NGCO companies.

Each sample subset was partitioned into two groups: (1) those companies that received GCOs and were predicted as bankrupt (GCO/bankrupt group) and, (2) those that received NGCOs and were predicted as nonbankrupt (NGCO/nonbankrupt group). The auditors’ opinions were correlated with the models’ predictions when the auditors issued GCOs (NGCOs) and the models predicted companies as bankrupt (nonbankrupt).

The pre-B and post-B samples were used to evaluate the correlation between the X, Y, and Z-score models’ predictions and auditors’ opinions for bankrupt companies. Pre-SAS No. 59 studies indicated that models routinely outperformed auditors at signaling impending failures. For example, Altman (1982) indicated the Z-score model (auditors) provided early warning signals of subsequent failure in 86.2% (48.1%) of the bankrupt companies in his sample. However, Chen and Church (1992) suggested that, under the provisions of SAS No. 59, auditors may use a different process than was used previously in deciding whether to issue GCOs. Binomial tests were used to compare the GCO/bankrupt and NGCO/nonbankrupt groups using the pre-B sample to those using the post-B sample. This test evaluated whether auditors’ decisions in going concern evaluations for bankrupt companies were more consistent with the models’ predictions after the issuance of SAS No. 59.

The pre-FD and post-FD samples were used to evaluate the correlation between the X, Y, and Z-score models’ predictions and auditors’ opinions for financially distressed companies other than bankruptcies. As previously indicated, prior studies limited their samples to include only bankrupt companies; however, auditors must decide whether to issue GCOs to firms from a variety of financial distress situations, not just possible bankruptcies. Binomial tests were used to compare the GCO/bankrupt and NGCO/nonbankrupt groups using the pre-FD sample to those using the post-FD sample. This test evaluated whether auditors’ decisions in going concern evaluations for companies from various financial distress situations were more consistent with the models’ predictions after the issuance of SAS No. 59.

Table 3 reports the proportion of bankrupt companies that were predicted as bankruptcies and that also received GCOs (GCO/bankrupt firms). Using Altman’s (Zmijewski’s) model, 56.3% (54.3%) of the post-SAS No. 59 and 53.1% (52.8%) of the pre-SAS No. 59 bankrupt companies were GCO/bankrupt firms. The results of binomial tests indicated that the proportions of GCO/bankrupt companies were not significantly different between the pre and post-SAS No. 59 periods for both the Altman and Zmijewski models. Using Ohlson’s model, the proportion of post-SAS No. 59 bankrupt companies that were GCO/bankrupt firms (53.5%) was significantly higher than that of pre-SAS No. 59 bankrupt companies that were GCO/bankrupt firms (41.6%). The findings suggest that the models’ predictions and auditors’ opinions using bankrupt firms were not more consistent after the issuance of SAS No. 59 except for the Ohlson model’s predictions.

Table 3 also reports the consistency between the models’ predictions and auditors’ opinions using financially distressed firms other than bankruptcies (other distressed in Table 3). Using the X, Y, and Z-score models, 58.8%, 52.1%, and 65.8% of the post-SAS No. 59 other distressed companies were GCO/bankrupt firms. For the pre-SAS No. 59 other distressed companies, 46.5%, 42%, and 52.6% of the firms were GCO/bankrupt firms. Though the post-SAS No. 59 proportions of GCO/bankrupt firms were higher than those of the pre-SAS No. 59 samples, binomial tests indicated that the proportions were not significantly different between the pre and post-SAS No. 59 samples. That is, SAS No. 59 did not affect the consistency between models’ predictions and auditors’ opinions using financially distressed companies other than bankruptcies.

The proportions of bankrupt companies predicted as nonbankruptcies that received NGCOs (NGCO/nonbankrupt firms) also are reported in Table 3. For the Altman and Zmijewski models, the proportions of post-SAS No. 59 bankrupt firms that were NGCO/nonbankrupt firms (63% and 52.1%) were significantly lower than those (82.4% and 76.5%) of the pre-SAS No. 59 bankrupt companies. Using Zmijewski’s model, the proportion of post-SAS No. 59 other distressed companies that were NGCO/nonbankrupt firms (62.1%) also was significantly lower than that (85%) of the pre-SAS No. 59 other distressed firms. The findings related to the proportions of NGCO/bankrupt firms were consistent with those for GCO/bankrupt firms discussed above. Though the changes in the proportions of GCO/bankrupt firms were not significant for all models (for the both bankruptcies and other distressed companies), the proportions of post-SAS No. 59 GCO/bankrupt firms were always higher than those of pre-SAS No. 59 GCO/bankrupt firms; consequently, the proportions of post-SAS No. 59 NGCO/nonbankrupt firms were always lower than those of pre-SAS No. 59 NGCO/nonbankrupt firms.

The consistency between the Altman, Zmijewski, and Ohlson models’ predictions and auditors’ opinions before and after the issuance of SAS No. 59 was evaluated in this study's tests. It is important to note that prior to the issuance of SAS No. 59, auditors had a passive role in evaluating the going concern question. However, SAS No. 59 significantly increased the auditors' responsibilities by requiring auditors to take an active role in evaluating a firm's ability to continue as a going concern.

The tests of this study provide insightful results related to findings of studies that used data from periods prior to the issuance of SAS No. 59. Pre-SAS No. 59 studies indicate that prediction models routinely outperformed auditors at signaling impending failures. The results of pre-SAS No. 59 studies suggest that prediction models could be useful in an auditor's evaluation of an entity's ability to continue as a going concern. The findings of this study generally indicate that the consistency between auditors’ opinions and models’ predictions did not change after the issuance of SAS No. 59. Consequently, it seems that the usefulness of prediction models in the going concern evaluation was unchanged after the issuance of SAS No. 59. This finding suggests that prediction models continue to outperform auditors at signaling impending failures in post-SAS No. 59 periods. As a result, prediction models continue to be valuable analytical tools that can assist auditors in the going concern evaluation.

This study also provides insights related to the impact that SAS No. 59 had on auditors' going concern evaluations. Prior research suggests that, under the provisions of SAS No. 59, auditors may use a different process than was used previously since the standard increased auditors’ responsibilities for going concern evaluations. In addition, prior studies suggest that SAS No. 59 likely increased auditors' costs of issuing "wrong" audit opinions. For example, auditors' costs of issuing NGCOs to firms that failed may increase after the issuance of SAS No. 59. The suggestion that SAS No. 59 changed the performance of auditors' opinions at signaling impending failure or that the standard changed the process auditors use in evaluating the going concern assumption presumes that SAS No. 59 altered existing practice as it relates to going concern evaluations.

Essentially, the findings of this study indicate that the increased responsibilities required by SAS No. 59 did not affect auditors' decisions in evaluating the going concern issue. It appears that auditors had assumed the "more responsible" role in going concern evaluations prior to the issuance of SAS No. 59 and that the ASB promulgated what actually was occurring in practice. Perhaps practitioners made "conservative" decisions when evaluating an entity's ability to continue as a going concern in pre-SAS No. 59 periods. In sum, the evidence from this study indicates that the ASB simply codified existing practice with the issuance of SAS No. 59.

AICPA. 1981. The Auditor’s Consideration When a Question Arises About an Entity’s Continued Existence. SAS No. 34. (March). New York: AICPA.

_____. 1988a. Analytical Procedures. SAS No. 56. (April). New York: AICPA.

_____. 1988b. The Auditor’s Consideration of an Entity’s Ability to Continue as a Going Concern. SAS No. 59. (April). New York: AICPA.

Altman, E. I. 1968. Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy. The Journal of Finance (September): 589-609.

_____. 1982. Accounting Implications of Failure Prediction Models. Journal of Accounting, Auditing, and Finance (Fall): 4-19.

_____. 1993. Corporate Financial Distress and Bankruptcy. 2nd ed. John Wiley & Sons, Inc.: New York.

_____, and McGough, T. 1974. Evaluation of a Company as a Going Concern. Journal of Accountancy (December): 51-57.

Barth, M. E., Beaver, W. H., and Landsman, W. R. "Valuation Characteristics of Equity Book Value and Net Income: Tests of the Abandonment Option Hypothesis." Working Paper, Stanford University, Stanford.

Blocher, E. and Loebbecke, J. K. 1993. Research in Analytical Procedures: Implications for Establishing and Implementing Auditing Standards. The Expectation Gap Standards: Progress, Implementation Issues, and Research Opportunities. Jersey City: AICPA.

Carcello, J. V., Hermanson, D. R., and Huss, H. F. 1995. Temporal Changes in

Bankruptcy-Related Reporting. Auditing: A Journal of Practice & Theory (Fall): 131-143.

Carmichael, D. R. and Pany, K. 1993. Reporting on Uncertainties, Including Going Concern. The Expectation Gap Standards: Progress, Implementation Issues, and Research Opportunities. Jersey City: AICPA.

Chen, K. C. W. and Church, B. K. 1992. Default on Debt Obligations and the Issuance of

Going Concern Opinions. Auditing: A Journal of Practice & Theory (Fall): 30-49.

Chen, C. W. and Wei K. C. 1993. "Creditors’ Decisions to Waive Violations of Accounting-Based Debt Covenants." The Accounting Review (April): 218-232.

Commission on Auditors’ Responsibilities, Cohen, M. F. 1978. Report Conclusions and Recommendations:New York.

Deakin, E.B. 1977. Business Failure Prediction: An Empirical Analysis. In Financial Crisis: Institutions and Markets in a Fragile Environment, edited by E. I. Altman and A. W. Sametz, New York: John Wiley & Sons: 72-98.

Dugan, M. and Zavgren, C. 1988. "Bankruptcy Prediction Research: A Valuable Instructional Tool." Issues in Accounting Education (Spring): 48-64.

Johnson, V. E. and Khurana, I. K. 1995. Auditor Reporting for Bankrupt Companies: Evidence on the Impact of SAS No. 59. Research in Accounting Regulation (vol. 9): 3-22.

Jones, F. L. 1987. Current Techniques in Bankruptcy Prediction. Journal of Accounting Literature. (vol. 6): 131-164.

Koh, H. C. 1991. Model Predictions and Auditor Assessments of Going Concern Status. Accounting and Business Research (Vol. 21): 331-338.

LaSalle, R.E. and Anandaragan, A. 1996. Audiotrs’s Views on the Type of Audit Report Issued to Entities with Going Concern Uncertainties. Accounting Horizons (June): 51-72.

Levitan, A. S. and Knoblett, J. A. 1985. Indicators of Exceptions to the Going Concern Assumption. Auditing: A Journal of Practice and Theory (Fall): 26-39.

Mckee, T. E. 1989. Modern Analytical Auditing. Greenwood Press, Inc.: Connecticut.

Ohlson, J. A. 1980. "Financial Ratios and the Probabilistic Prediction of Bankruptcy." Journal of Accounting Research (Spring): 109-131.

Ott, R. L. 1991. An Introduction to Statistical Methods and Data Analysis. 4th ed, Wadsworth, Publishing Inc.: California.

Paquette, L. R. and Skender, C. J. 1996. Using a Bankruptcy Model in the Auditing Course: The Evaluation of a Company as a Going Concern. Journal of Accounting Education Vol. 14 No. 3: 319-329.

Raghunandan, K. and Rama, D. V. 1995. Audit Reports for Companies in Financial Distress: Before and After SAS No. 59. Auditing: A Journal of Practice and Theory (Spring): 50-63.

Zavgren, C. 1983. "The Prediction of Corporate Failure: The State of the Art." Journal of Accounting Literature (vol.2): 1-38.

Zmijewski, M. E. 1984. Methodological Issues Related to the Estimation of Financial Distress Prediction Models. Journal of Accounting Research 24 (Supplement): 59-82.