Experimental Investigation of the Effects of Cognitive Elaboration on Accounting Learning Outcomes

This study analyzed how self-generated elaboration, instructor-assisted elaboration, and self-generated followed by instructor-assisted elaboration, affect accounting students’ acquisition of procedural knowledge, intellectual skills, and their attitudes towards learning. The results indicate that the self-generated elaboration instructional strategy improves accounting students’ acquisition of procedural knowledge and intellectual skills more than the instructor-assisted elaborations. However, the effects of these instructional strategies on attitudes are not significantly different. Also, our three-way ANOVA results indicate that the students who have higher GPAs and pretest scores tend to have more intellectual skills, but not necessarily more procedural knowledge or more positive learning attitudes, using the self-generated elaboration instructional strategy.


Introduction
The development of accounting expertise involves the acquisition of various sets of knowledge and skills.Given the breadth of the knowledge and skills required in the accounting profession it may be difficult to develop expertise in school.The ability to learn and apply knowledge to different situations is therefore very critical to accounting students.Many different expert-learning instructional strategies have been explored to help students learn how to learn (Stone, Shelley, & Pincus 1997, Cook & Hazelwood, 2002;Benek-Rivera & Mathews, 2004).Cognitive elaboration approaches emphasize active learning (Anderson, 1993) and have therefore received considerable attention as an alternative mode of instruction.However, little attention has been paid to the effects of cognitive elaboration approaches on accounting learning outcomes.
This study investigates the effects of three alternative forms of cognitive elaboration on first accounting course students' learning outcomes.Specifically, this study examines how self-generated elaboration, instructor-assisted elaboration, and self-generated followed by instructor-assisted elaboration, affect first accounting course students' acquisition of procedural knowledge, intellectual skills, and students' attitudes.

Motivation of the Study
This research is motivated by two issues.First, this study explores alternative instructional models in the field of accounting.Traditional instructional models were described as instructor-centred: simply presenting information, giving examples, and providing opportunities for practice (Hosal-Akman et al., 2010).In complex and ambiguous settings such as accounting, a simple presentation of information is not adequate for students to learn (Bonner, 1999;Cook & Hazelwood, 2002).Especially the first accounting course is a gateway subject for business education.Students' experience with their first accounting course is an important factor in their selection of accounting as a major (Geiger & Ogilby, 2000).Therefore, find an optimal teaching strategy for introductory accounting course is a central recurring theme in business education (Lloyd & Abbey, 2009) In addition, the Accounting Education Change Commission states that students must become active participants in the learning process and instructors need to utilize methods to promote interaction with students (AECC, 1990).Teaching students the ability of learning to learn has therefore become an important task for accounting educators today (Lightbody, 1997;Cook & Hazelwood, 2002).This study attempts to provide some additional evidence on the effects of active learning in the accounting learning process.
Secondly, this research explores how different cognitive elaboration strategies might help teachers achieve desired learning outcomes for their first accounting course students.As previously stated, little is known about the potential effects of cognitive elaboration strategies on accounting students' learning outcomes.This study seeks to provide some insights in that direction.

Cognitive Elaboration
Cognitive elaboration refers to the generation and integration of new information into one's pre-existing knowledge structure (Anderson, 1983).In the context of learning, cognitive elaboration refers to an instructional strategy that helps students to elaborate on new course material in order to relate it to other information in the course material and/or information that the students already possess (Shuell, 1988).There are three alternative forms of cognitive elaboration, including the self-generated elaboration, the instructor-assisted elaboration, and the self-generated elaboration followed by the instructor-assisted elaboration.The self-generated elaboration encourages students to take a central role and generate their own way of integrating new course material with pre-existing knowledge structures.On the other hand, both the instructor-assisted elaboration and the combined elaboration models allow students to take a peripheral route by giving them instructions to generate the integration process without active thinking.

Learning Outcomes
There are several taxonomies of learning outcomes available in educational and cognitive psychology (Bloom, Engelhart, Furst, Hill, & Krathwohl 1956;Anderson, 1976;Gagne, 1984;Stone et al., 1997).Three important categories of learning outcomes documented in educating for expertise are procedural knowledge, intellectual skills, and attitudes (Stone et al., 1997).Declarative and intellectual skills are two long-term stores of knowledge.
Procedural knowledge consists of facts, instructions, examples, and definitions (Anderson, 1993).It is factual knowledge that students are able to recall in essentially the same form as it was originally presented to them (Bonner, 1999).Students are not asked to apply knowledge but merely to retrieve from memory.
Intellectual skills are the application of knowledge to a specific use or task (Anderson, 1993).It includes knowledge of concepts, rules, and procedures needed for problem solving in a given domain.The key challenge with regard to intellectual skills is for students to be able to apply these skills to novel situations (Bonner, 1999).Intellectual skills cannot be verbalized, but are manifested through performance on a task.
Attitudes are beliefs that influence individuals' choices and actions (Stone et al., 1997).Attitudes determine to what extent the students' recognize that learning is needed (Feldman, 1986).Gal and Ginsburg (1994) find negative attitude impede learning or hinder the extent to which students will develop useful information and apply what they have learned outside the classroom.They believe that one goal of education is to engender in students a positive outlook about learning and its applications, and have confidence in themselves in the learning process.

The Impact of Cognitive Elaboration on Learning Outcomes
Cognitive research emphasizes the cognitive processes that turn novices into experts (Anderson, 1983).Prior research suggests that different categories of learning outcomes require different instructional approaches and make different information processing demands for learning (Anderson, 1985;Gagne, 1984).Traditional accounting education has emphasized procedural knowledge (Bonner & Walker 1994).Other research has focused on the use of different instructional strategies to improve the acquisition of procedural knowledge in the first accounting course (Gieger &Ogilby 2000; Lloyd & Abbey 2009).Schadewald and Limberg (1990) investigated the effects of cognitive elaboration on students' ability to memorize facts related to tax accounting.Their results show that students under self-generated elaboration have greater ability to recall tax facts than students under instructor-assisted elaborations.Hermanson (1994) found that students who generated their own elaboration had better recall of both simple and complex accounting and tax concepts than did students who were provided with the instructor-generated elaborations.
The public accounting environment is changing rapidly, requiring an increased emphasis on complex decision-making and problem solving skills.Accountants rely more heavily on analytical skills and the ability to transfer knowledge than ever before (Elliott, 1998).In order to improve students' abilities in acquisition and transfer of knowledge, a focus on intellectual skills and attitudes is necessary.With intellectual skills, students are able to differentiate and integrate alternative problem solving perspectives, identify accounting-related information resources, structure problem solutions, and develop written communication skills (Stone et al., 1997).Attitudes are also very important as they strongly influence the recognition of the need for learning procedural knowledge and the need for applying intellectual skills (Ennis, 1987).Similar experience is found in other learning fields such as Statistics (Gal & Ginsburg. 1994).Gal and Ginsburg (1994) shows that teachers should build an emotionally supportive atmosphere in a problem-solving environment for learning statistics, so students feel safe to explore, conjecture, hypothesize, and brainstorm; are motivated to keep working on problems which may not have right or wrong solutions and may require extended investment of energy; feel comfortable with temporary confusion or a state of inconclusive results; and are not afraid to experiment with applying different (statistical) tools or methods.
Certain forms of cognitive elaboration encourage the development of intellectual skills and attitudes by emphasizing active learning (Anderson, 1985;Feldman, 1986;Stone et al., 1997).The major objective of this study is to explore whether teachers could help students attain desired learning outcomes by choosing different cognitive elaboration strategies.To achieve this goal, this study attempts to provide experimental evidence on the effects of alternative forms of cognitive elaboration on the accounting students' learning outcomes.

Hypotheses Development
The self-generated elaboration instructional model encourages students to be actively involved in the learning process.The intent of this strategy is to improve students' procedural knowledge by allowing students to independently acquire it.Stone et al. (1997) compared the effectiveness of a comprehensive instructional program, Project Discovery (PD), with traditional accounting education on students' learning outcomes.Their results show that PD effectively emphasizes acquiring intellectual skills and improving attitudes without losing traditional accounting procedural knowledge.Therefore the self-generated elaboration might more effectively increase students' procedural knowledge than the instructor-assisted models by developing students' abilities to independently acquire the knowledge.However, prior research also shows that instructional strategies that emphasize active learning de-emphasize declarative, but rather enhance students' ability to develop intellectual skills and attitudes (Anderson, 1993).Students under the self-generated elaboration instructional model might exhibit less procedural knowledge than students under the instructor-assisted models.Since the research is contradicting, our first hypothesis will be as follows: Hypothesis 1: Students who are taught under self-generated elaboration will exhibit no difference in procedural knowledge than students who are taught under instructor-assisted elaborations.Developing and increasing cognitive complexity is an important objective of instructional strategies.Cognitive complexity refers to the ability to differentiate alternative perspectives and integrate these perspectives into a well-reasoned analysis (Streufert & Swezey 1986).In complex, ambiguous environments such as accounting, increasing cognitive complexity is necessary for improving problem-solving skills (Breuer & Tennyson 1995).The self-generated instructional strategy seeks to enhance students' cognitive complexity by letting students take a central role in the learning process.Consequently, hypothesis 2 is stated as follows: Hypothesis 2: Students who are taught under the self-generated elaboration exhibit more intellectual skills than students who are taught under the instructor-assisted elaborations.
Different cognitive elaborations have different impacts on student's attitudes toward instruction.The self-generated instructional strategy involves students in the learning process and elaborates on their cognitive effort (Pincus, 1991).Therefore, hypothesis 3 states as follows: Hypothesis 3: Students who are taught under the self-generated elaboration exhibit more positive attitudes about their learning method than students who are taught under the instructor-assisted elaborations.
Extensive research indicates that individual ability has significant impact on accounting performance and GPA has been found to be one of the strongest predictors of accounting performance ( (Hite & Parry, 1994;Hill, 1998).Therefore, GPA was included as a control variable in the statistical analyses to control for performance difference that result from individual differences in ability.

Subjects
Seventy-seven students in an introductory accounting course were used in this experiment.Twenty-six students received the self-generated elaboration treatment, twenty-six received the instructor-assisted treatments, and the other twenty-five received the self-generated followed by the instructor-assisted treatment.

Procedures
The experiment was carried out in one class session.In order to determine the initial knowledge of the subject, a pretest/posttest design was used.Table 1 presents the experimental design for this study, including tasks and conditions for the pretest, the treatments, and the posttest.Students were first given five minutes to respond to the pretest, which included definitions and conceptual questions, case problems and a simple questionnaire.Then students were randomly chosen for the treatments (Cognitive elaborations materials).Subjects were given ten minutes to read and engage those elaborations.Finally, students were administered the posttest, which included the procedural knowledge test (definition and conceptual questions), the intellectual skills test (Case analysis), and the attitudes test (questionnaire).Students were instructed not to discuss the experiment with classmates and that the experiment was a class exercise.

Materials/ Strategies
The three cognitive elaboration strategies were designed using similar procedures to those used by Schadewald and Limberg (1990) and Choo and Tan (1995).The actual accounting problems include steps to calculate the operating cycle for a company.The instruction was different for each cognitive elaboration strategy.A pilot test was conducted to get feedback on the experimental materials before they were given to the students.The pilot test was administered to five students from the testing class.We wanted to know whether students would understand the questions on the experimental materials and whether there are ambiguous questions.The five students had no problem understanding the questions, so the same experimental materials were used in the test.

The Self-generated elaboration:
The self-generated elaboration required the students to follow the instructions given and helpful hints to figure out their own solutions to the operating cycle problems.Lastly the students were asked to verify their answers with the answer keys.Students under this elaboration were guided through the self-cognitive elaboration process.

The Instructor-assisted elaboration:
The instructor-assisted elaboration was operationlized by providing students with the suggested solutions to the problems.The students were prompted to follow the instructor-assisted elaboration in the traditional sense.That is, students were simply given the solution without explicitly guiding them through the cognitive elaboration process.
The Self-generated elaboration followed by the instructor-assisted elaboration: This combined elaboration was operated by first giving the students the above self-generated elaboration instructional material, and then providing them with the suggested solutions.In this process, the students first experienced a self-cognitive elaboration process, and were then given instructional help.To be consistent, all these elaboration strategies were administered in ten minutes.

Measures
The number of correct answers out of eight total questions on the pretest and posttest was used as measures for pretest and posttest performances.The pretest and posttest were graded by an independent person who had no knowledge of what was being tested.
Following Stone et al. (1997), we measure procedural knowledge by the definitions and concept questions adopted from the CPA exams.Intellectual skills were measured by the case questions.Attitudes were measured by asking the students to rank the instructional method on a seven-point scale.The GPAs were obtained from the pretest.

Results
None of the students knew the answers about the operating cycle during the pretest.Also, no significant difference was found in the average GPAs of the students in the three elaboration groups due to the random process of the subject selection.
Figure 1 presents the mean scores of correct answers for procedural knowledge and intellectual skill under the three cognitive elaboration strategies in the posttest.According to Figure 1, students taught under the self-elaboration instructional strategy outperformed those taught under the instructor-assisted and the combined strategies on both the procedural knowledge and the intellectual skills tests.An analysis of variance showing the effects of all of the cognitive elaboration strategies can be found in Table 2.
The effect of the cognitive elaboration strategies is significant at the 10% level for the posttest (F = 2.5, P= 0.09).
The pretest is an insignificant covariate for the posttest performance (F= 0.24, p=0.63).The pretest score is merely a control variable, the previous knowledge makes no prediction of the effect of the elaboration strategies on the posttest performance.Table 3 presents the mean scores, standard deviations, t-test, and p-values for the differences among the three groups for procedural knowledge, intellectual skills, attitudes, and the total posttest scores.Hypotheses 1 predicts that there is no difference on procedural knowledge among the three cognitive elaborations, while Hypotheses 2 and 3 predict that students under the self-generated elaboration will outperform students under the other two elaboration strategies in terms of intellectual skills and attitudes.We find that Hypothesis 1 is partially supported.(self v. instructor: t=1.99, p=0.05; self vs. combination: t=0.40 p=0.69; instructor vs. combination: t=-1.41 p=0.17).That is, the self-elaboration strategy does improve students' Procedural knowledge at 5% significance level compared to the instructor-assisted elaboration strategy.The results are positive but are not statistically significant when comparing the self elaboration with the combined strategy.Hypothesis 2 is partially supported as well (self v. instructor: t=1.91, p=0.06; self vs. combination: t=0.67 p=0.51; instructor vs. combination: t=-1.05p=0.30).That is, the students taught under the self-elaboration strategy outperformed the students taught under the instructor-assisted strategy on intellectual skills at 6% significance level.However, the difference is not significant between the self-elaboration strategy and the combined strategy.e Hypothesis 3 is not supported (self v. instructor: t=-0.45,p=0.65; self vs. combination: t=-0.99 p=0.33; instructor vs. combination: t=-0.47 p=0.67), indicating the self-generated elaboration does not improve the students' attitudes compared to the other two strategies even though it improves students' procedural knowledge and intellectual skills.Hypothesis 4 predicts that there is a significantly positive relationship between the GPA and the learning outcomes.An analysis of variance in Table 4 presents the effects of the GPA on the posttest, procedural knowledge, intellectual skills, and attitudes.Panel A and Panel B show that the GPAs are significantly associated with the posttest scores and procedural knowledge at the 5% significance level (A: F=6.62, p=0.01;B: F=4.8, p=0.03).Panel C shows that the GPAs have significant effects on intellectual skills at the 10% level (F = 3.67, p=0.05).Panel D shows that there is an insignificant relationship between the GPA and the attitudes (F=2.3,p= 0.13), indicating that the students who have higher GPAs do not necessarily have more positive attitudes toward learning than those who have lower GPAs.Hypothesis 4 is partially supported.In addition, Table 4 presents an analysis of variance on the effects of cognitive elaboration strategies on the posttest, Procedural knowledge, intellectual skills, and attitudes.Similar to the results in Table 2, Panel A show that Cognitive elaboration strategies have significant effects on posttest at 5% significance level (F=2.96,P=0.05).However, no significant results were found for procedural knowledge, intellectual skills and attitudes.

Discussion
This study investigates the effect of one type of expert learning and problem solving strategy, i.e., cognitive elaborations, on accounting students' learning outcomes.The results show that the students taught under the self-generated elaboration instructional strategy perform significantly better on procedural knowledge and intellectual skills than the students under the instructor-assisted elaboration instructional strategy.Although the students under the self-elaboration strategy perform better than the students under the combined strategy, the differences are not significant.What is noticeable here is that the means of the two methods that involve self-generation seem to be much closer to each other than were the two instructor-assisted methods for both learning outcomes.According to these results, in order to promote knowledge transfer and development of accounting expertise, instructional strategies should be chosen that promote self-generation elaboration.
In addition, there is no significant difference of students' attitudes among the three forms of cognitive elaboration approaches.While there are no significant differences among the means, it was interesting that the mean for the self-generated method was lower than for the instructor assisted methods, but the highest mean was for the combined method.
Finally, the study finds that GPAs have a significant effect on students' acquisition of procedural knowledge, intellectual skills, but have an insignificant effect on students' attitudes towards learning methods.
Overall, the findings imply different forms of cognitive elaboration strategies do have different effects on the learning outcomes.To promote knowledge transfer and development of accounting expertise, instructional strategies should be chosen utilizing methods involving self-generated elaborations.
One of the major limitations of this study is its failure to capture a real cognitive elaboration instructional setting.
To avoid instructor effects and enhance the study's internal validity, the cognitive elaboration instructional strategies were conducted through instructional materials instead of real lectures.Future research could be conducted by giving real lectures to different classes to investigate the effects of cognitive elaboration on learning outcomes.Another limitation of this study is its weakened external validity.First, the sample size is too small; second, the sample was drawn from one class.Generalizations from this study are limited.

Figure 1 .
Figure 1.Effect of Cognitive elaborations on learning outcomes

Table 1 .
Experimental design

Table 2 .
The effect of cognitive elaborations and pretest on posttest, procedural knowledge, intellectual skills,

Table 3 .
The effect of cognitive elaborations on procedural knowledge, intellectual skills, attitudes, and posttest:

Table 4 .
The effects of cognitive elaborations and GPA on posttest, procedural knowledge, intellectual skills, and Table5presents the interactive effects of cognitive elaboration strategies, pretest, and GPA on posttest, procedural knowledge, intellectual skills, and attitudes.Panel A and Panel B show that there is no significant joint impact of cognitive elaboration strategies, pretest, and GPA on posttest (F=1.22,p= 0.38) or procedural knowledge (F=0.72,p=0.80).Panel C shows that the cognitive elaboration strategies, pretest, and GPAs are jointly and significantly affect the intellectual skills at the 5% significance level (F=2.45,p=0.05).Panel D shows that there is no joint effect of the cognitive elaboration strategies, pretest, and GPA on attitudes (F=0.57,p= 0.92).Thus, the results indicate that the students who have higher GPAs, pretest scores, self-generated elaboration instructional strategy would have more intellectual skills, but not necessarily more procedural knowledge or more positive learning attitude, than those who have lower GPAs, pretest scores, and other two cognitive elaborations.This suggests that the students' natural intelligence GPAs and pretest scores could influence the impact of cognitive elaborations on intellectual skills.Hypothesis 2 is again supported.

Table 5 .
The interactive effects of cognitive elaborations, GPA, and pretest on posttest, declarative, intellectual skills, and attitudes: Three-way ANOVA