EXAMINING BELIEF ADJUSTMENT MODEL AND INVESTORS OVERCONFIDENCE ON INVESTMENT DECISION MAKING - Perbanas Institutional Repository
EXAMINING BELIEF ADJUSTMENT MODEL AND INVESTORS
OVERCONFIDENCE ON INVESTMENT DECISION MAKING
R E S E A R C H A R T I C L E
Submitted to Qualify the Requirement of Undergraduate Program of
Accounting Department
Written by :
DYAH ERAS MITA
2013311078
SEKOLAH TINGGI ILMU EKONOMI PERBANAS
S U R A B A Y A
2017
EXAMINING BELIEF ADJUSTMENT MODEL AND INVESTORS
OVERCONFIDENCE ON INVESTMENT DECISION MAKING
Dyah Eras Mita
Accounting Undergraduate STIE Perbanas Surabaya
Email: Luciana Spica Almilia
STIE Perbanas Surabaya Email:
Jl. Nginden Semolo 34-36 Surabaya
ABSTRACT
This research aims to examine whether there is a different judgment between the investor who receives good news followed by bad news and the one who receives bad news followed by good news information order in the Step by Step and the End of Sequence disclosure pattern by using accounting information type and non-accounting information type and overconfidence characteristics on investment decision making. This research is included to the experimental design by using a mixed design of between-subjects and within-subject design and classified as an experimental research which uses 2x2x2x2 method. To find the result, it conducts normality test, independent sample t-test, and Mann-Whitney U test using SPSS version 22. Participants used in this research are undergraduate business students in STIE Perbanas Surabaya who are studying and/or has completed investment management and/or financial statement analysis courses who will serve as non-professional investors. The results obtained in this research showed that recency effect occurred between the investor who receives good news followed by bad news and the one who receives bad news followed by good news in the step by step disclosure pattern, while there is no order effect occurred when the disclosure pattern used is the end of sequence.
Key words: Step by Step, End of Sequence, Investment Decision, Overconfidence
INTRODUCTION
The development of investment in which had presented its annual reports in Indonesia is experiencing favorable 2013. While in 2014 there were 493 conditions that it becomes a good companies which had presented its condition for investors to invest in the financial statements and 494 companies stock market. The investors need which had presented its annual reports, and information to analyze their investment. It there were 486 companies which had means that there is accounting information presented its financial statements and 487 and non-accounting information needed companies which had presented its annual for investment decisions. The publication reports in 2015. This indicates an increase of financial statements and annual reports for company’s disclosure of the audited in the IDX website shows that there were financial statements and the audited annual 483 companies which had presented its reports. financial statements and 490 companies The factors that affect the model of the belief adjustment process. investment decisions can be described in a Hogarth and Einhorn (1992) proposed the belief adjustment model which proposed that individual who processes information sequentially will use the anchoring process and adjustment process. Belief adjustment model predicts that there is no order effect on the consistent evidence that occurs when an individual gets various evidence. The order effect will often come up when the disclosure pattern is sequential or step by step (SbS).
Research conducted by Luciana Spica Almilia and Supriyadi (2013) concluded that there is a difference on investment decision among participants who received good news followed by bad news compared to participants who received bad news followed by good news in sequentially information disclosure. participants who received good news followed by bad news compared to participants who received bad news followed by good news in simultaneously information disclosure.
The understanding of investors about those informations or any disclosure presented by the company is important to make investment decision because it reflects uncertainty faced by the company. Ghosh and Wu (2012) explained that the measurement of financial and non- financial performance and their favorableness have interactive impact on the analyst recommendations. This phenomenon is encouraging researcher to examine the effect of accounting information and non-accounting information on investment decision making. This study also tries to understand the attitudes of investors influenced by errors in judgments or even mental routines. Trivers (2004) stated that in self- deception theory, individuals are designed to think they are better (smarter, stronger) than they really are. Self-deception can explain overconfidence, the tendency of decision makers to give excessive weight to the assessment of knowledge and accuracy of information possessed and ignore the public information available. Indeed, investors are not fully rational and their demand for financial assets is often affected by their beliefs or feelings which are not clearly justified by economic fundamentals.
Based on the background above, this study is aimed to examine the information order (++-- and --++), disclosure pattern (step-by-step and end- of-sequence), information type (accounting information and non-accounting information) and identifying investors overconfidence on investment decision making. According to the explanation above, the author proposes title
“Examining Belief Adjustment Model and Investors Overconfidence on Investment Decision Making” to
headline the purpose of study. This is especially to identify the effect of belief adjustment model and investor overconfidence towards investment decision making.
THEORETICAL FRAMEWORK AND HYPOTHESES Belief Adjustment Theory
Hogarth and Einhorn (1992) proposed the belief adjustment theory which predicts order effects in almost all cases of response mode (SbS or EoS), task complexity, and length of information. This theory proposed that an individual who processes information sequentially will use the anchoring process and adjustment. Specifically, the belief adjustment model predicts that there are not any order effects on that consistent evidence (totally negative or reversed, but the recency order effect occurs when an individual acquires various evidence (some are negative and some are positive). The primary advantage of the belief adjustment model developed by Hogarth and Einhorn (1992) is the inclusion of three main characteristics of evidence employed in Bayes Theorem (direction, strength and type) while it also broadened Bayes Theorem by including two additional characteristics that are ignored by Bayes Theorem, which are information order and information presentation model.
The direction of evidence shall indicate whether the evidence encourages or discourages the individual beliefs. The additional evidence that encourages the belief is called good news or positive
evidence (e.g. net income of the company
increases than the last period, earnings per share of the company increases than the last period), while the additional evidence that discourages belief is called bad news or negative evidence (e.g. net income of the company decreases than the last period, earnings per share of the company decreases than the last period). The second strength or level of evidence that courage or discourage the belief. Type of evidence can be categorized into consistent evidence or mixed evidence. When all additional evidence has the same direction (both positive and negative), it is categorized as
the consistent type of evidence , while some
evidence is negative and some are positive, it is categorized as mixed evidence. The information presentation models in the belief adjustment theory proposed by Hogarth and Einhorn are a step by step (SbS) and the end of sequence (EoS). The step by step (SbS) is a pattern of information presentation when investors had the transaction based on simple shared information (e.g., financial statements and the quarterly interim non-accounting information obtained from the mass media) and performed in sequence. The end of sequence (EoS) is a pattern of information presentation when investors had the transaction based on complete information and the whole reports obtained at such particular timelines (e.g., complete annual report which does not only contain financial statement).
The Information Order Effect and Disclosure Pattern
The order effect occurs when the decision made by individual is different after receiving evidences in different order. The character of evidence is mixed between positive information and negative information. When the first information in the order has a large effect on the individual belief, then such order effect is called
primacy effect . On the other hand, when
the last information is the one that gives the largest effect, it is called recency effect. The results of prior studies (Ashton and Ashton, 1998; Tubbs et al., 1993) indicate that the recency effect is not found when the acquired evidences give consistent information, both positive information and negative information. The recency effect occurs when the evaluated evidences have mixed information that is positive information and negative information.
The phenomena of order effect happen when judging new evidence and the next adjustment is based on insufficient additional evidence. Hogarth and Einhorn (1992) proposed the belief adjustment model that predicts the review when an individual evaluates the complex short serial evidence and mixed evidence. It is called short series evidence when there are maximally 12 items of evidence. It can be concluded that it is called long series evidence when there are more than 17 items of evidence. The complexity is due to the length of evidence item. Mixed evidences consist of both positive and negative items. The belief adjustment model predicts the order effect on all cases of respond model (SbS or EoS), job complexity and information length. Below is the expected order effect based on The Belief Adjustment Model proposed by Hogarth and Einhorn (1992)
This research employs simple information type, long series information by using Step by Step and End of Sequence disclosure pattern, the information order
No Effect Primacy
Self-Deception Theory (Trivers, 2004) predicts that when a person is unconsciously believing themselves has the ability above average and individuals are designed to think they are better (smarter, stronger) than they really are. Thus, an individual may be able to posture more effectively to potential competitors by genuinely believing that his abilities are greater than they are. Self-deception can explain overconfidence (a tendency to overestimate one’s ability or judgment accuracy), and dynamic processes that support overconfidence such as biased self-attribution (a tendency to attribute successes to one’s own ability and failure to bad luck or other factors). Thus, it can be concluded that overconfidence behavior may be caused by the belief of high overconfidence investors that they have ability or more experience and specialized knowledge or more knowledge in the field of stock market investments.
Self-Deception Theory
lead to poor investment performance. The research done by Benos (1998) using the auction market research, concludes that overestimating the accuracy of the information will lead to increase trading volume of investors.
The theory of excessive trading argues that overconfidence behavior will lead to the tendency of investors to practice aggressive and excessive trading strategies (e.g. high trading volume and/or high trading frequency). Ult imately, it will
that is the most common characteristics found in humans that reflect one's tendency to overestimate the ability. Overconfidence is a key concept to understand why investment strategies are so actively pursued and trading is excessive. According to Klayman et al. (1999) and Kufepaksi (2007), a person ’s overconfidence level can be identified through calibration test of confidence level. Calibration test is a procedure to test and identify the combination of the level of knowledge and level of confidence that shape one's level of overconfidence based on a specific questionnaire designed specifically for these purposes.
In the modern financial theory, investors are assumed to be rational in their efforts to identify and process the relevant information for optimal decisions. But in recent years, emerged a variety of empirical evidence showing the existence of investor behavior that deviates from these predictions. The excessive trading theory can explain overconfidence behavior,
Excessive Trading Theory
Note: Yellow colored column shows the main focus of this current research
No Effect Primacy
No Effect Primacy
Primacy Primacy
Table 1: Expected Order Effect based on Belief Adjustment Model Simple Complex
Consistent Information Set (++++ or ----) Short Long
Recency Primacy
Recency Primacy
Recency Primacy
Primacy Primacy
Mixed Information Set (++-- or --++) Short Long
Step by Step
End of Sequence
Step by Step
End of Sequence
- (good news followed by bad news) or
- (bad news followed by good news), the type of information is accounting information and non-accounting information, and investors overconfidence level. The research uses long series
Figure 1
Theoretical Framework
The previous theoretical framework describes that this study is aimed to identify the effect of belief adjustment model and investor overconfidence towards investment decision making. Issues related to the information disclosure pattern (SbS or EoS) which may affect investment decision making as described in the previous theoretical framework is supported by the result of prior studies which have shown greater belief revisions for sequentially disclosed information than for simultaneous disclosures (Hogarth and Einhorn,1992). Thus, to be conservative and in order to examine the existence of order effect and information disclosure pattern on investment decision making, the above discussion is the basis for the following hypotheses: H
1
: There is a different judgment between the investor who receives good news followed by bad news (++--) and the one who receives bad news followed by good news (--++) information order in the step by step disclosure pattern on accounting information type. H
2 : There is a different judgment
8 manipulated conditions of between subjects (two conditions of information order on the two conditions of information disclosure pattern by using the two conditions of the information type). Then, the participants are asked to repeat their experimental task two times. Within- subject design was used to compare the participants stock price judgment before and individual characteristics.
3. Scenario
II, in this scenario participants will be given bad news followed by good news (--++) information order by using Step by Step (SbS) disclosure pattern on accounting information.
2. Scenario
1. Scenario I, in this scenario participants will be given good news followed by bad news (++--) information order by using Step by Step (SbS) disclosure pattern on accounting information.
Experiment in this research is done by using paper and pencil test, it means that the participants are asked to answer the questionnaire manually. Participant in this research will be asked to examine the role of investor in valuing the company performance based on accounting information and non-accounting information given. This research contains of 8 scenarios. Those scenarios are described as follows:
Experimental Procedure
Experimental design used was a mixed design of between-subjects and within-subject design. Participants will be distributed to the two conditions of information order (good news followed by bad news or bad news followed by good news), two conditions of information disclosure pattern (Step by Step or End of Sequence), and the two conditions of the type of information (accounting information or non-accounting information). Thus, there are
between the investor who receives good news followed by bad news (++--) and the one who receives bad news followed by good news (--++) information order in the step by step disclosure pattern on non- accounting information type. H
This research is included to the experimental method. This research investigates the phenomenon by manipulating the circumstances or conditions through certain procedure, then examines the manipulation result and interprets it (Ertambang Nahartyo, 2012). This method is chosen because the experimental method is strong enough in the case of showing the causal relationship among research variables.
RESEARCH METHOD Experimental Design
between the investor who receives good news followed by bad news (++--) and the one who receives bad news followed by good news (--++) information order in the end of sequence disclosure pattern on non- accounting information type.
4 : There is a different judgment
between the investor who receives good news followed by bad news (++--) and the one who good news (--++) information order in the end of sequence disclosure pattern on accounting information type. H
3 : There is a different judgment
III, in this scenario participants will be given good news followed by bad news (++--) information order by using Step by Step (SbS) disclosure pattern on non-accounting information.
4. Scenario
IV, in this scenario participants will be given bad news followed by good news (--++) information order by using Step by Step (SbS) disclosure pattern on non-accounting information.
5. Scenario V, in this scenario participants will be given good news followed by bad news (++--) information order by using End of Sequence (EoS) disclosure pattern on accounting information.
6. Scenario VI, in this scenario participants will be given bad news followed by good news (--++) information order by using End of Sequence (EoS) disclosure pattern on accounting information. participants will be given good news followed by bad news (++--) information order by using End of Sequence (EoS) disclosure pattern on non-accounting information.
8. Scenario VIII, in this scenario participants will be given bad news followed by good news (--++) information order by using End of Sequence (EoS) disclosure pattern on non-accounting information.
First, participants are asked to answer experiment psychological question to measure investor overconfidence characteristics. Then, participants are asked to revalue PT. NYE share, which is a virtual corporation but the provided data is based on the real data taken from Indonesia Stock Exchange (BEI). Virtual corporation used in this research is a corporation in the field of Herbal Medicine and Supplement, Food and Beverage, and Pharmacy. This corporation was choosen because this virtual corporation can survive in all of economics condition in Indonesia. Moreover, PT. NYE share has been traded in Jakarta Stock Exchange formerly known as Indonesia Stock Exchange (BEI) nowadays.
Below is the company’s background given in this experimental research:
PT. NYE. was established since 1951. The main sector of the company is in Herbal Medicine and Supplement, Food and Beverage, and Pharmacy Industry.
PT. NYE’s products in Indonesia include some well-known brands, and not only
sold for domestic consumer but also imports its products to another country. PT. NYE. first offer the shares to the public in 2013 and listed on the Indonesia Stock Exchange since December 18, 2013.
IPO (Initial Public Offering) price of NYE Inc. is Rp. 580, - per share with the number of shares as much as 1.500.000.000 (one billion five hundred the total shares. The share price of PT. NYE at the beginning of 2015 was Rp 610, - per share as reference.
Then, participants are asked to revalue each type of information regarding investment provided by using Step by Step (SbS) and End of Sequence (EoS) disclosure pattern. After that, participants are asked to give the stock price recommendation based on the provided information, that is, company financial statements and non-accounting information disclosure taken from annual report or mass media with the multiple price of +100 for very good news information and -100 for very bad news information. Then, they will be asked to fill out the manipulation check question. Participants in the experiment are also asked to answer multiple choice question concerning their skill on investment management and financial statement analysis. Therefore, the experimental procedures for the step by step (SbS) disclosure pattern are executed as below:
1. Participants are asked to respond the experiment psychological question to measure investor overconfidence characteristics.
2. Reading company background.
3. Information containing beginning share price is given (by using Rp 610, - as the initial stock price).
4. Net Sales of the company increases than the last period.
15. Operating Income of the company decreases than the last period.
14. Current assets of the company decreases than the last period.
13. Net Sales of the company decreases than the last period.
12. Return on Equity ratio of the company decreases than the last period.
11. Return on Assets ratio of the company decreases than the last period.
10. Net Income of the company decreases than the last period.
9. Net Profit Margin of the company increases than the last period.
8. Total Assets of the company increases than the last period.
7. Earnings Per Share of the company increases than the last period.
6. Operating Income of the company increases than the last period.
5. Current assets of the company increases than the last period.
2. Return on Assets ratio of the company increases than the last period. company increases than the last period.
4. Giving accounting information (financial statements) and non- accounting information (annual report or information obtained from mass media) sequentially.
1. Net Income of the company increases than the last period.
Accounting Information Items:
7. Debriefing session Information items used in this experimental research are 18 items which are grouped into 9 (nine) good news items and 9 (nine) bad news items taken from company financial statement for accounting information items. Meanwhile, the non-accounting information items used in this experimental research is taken from annual report and information obtained from the mass media as much as 18 items which are categorized into 9 (nine) good news items and 9 (nine) bad news items.
6. Participants are asked to respond the manipulation check question, question to measure the participant basic understanding on financial statement analysis, investment management and company demographic items.
5. Revaluing company stock price that will be bought as much as one judgment for all informations given (accounting information and non- accounting information given).
4. Giving accounting information (financial statements) and non- accounting information (annual report or information obtained from mass media) once or simultaneously
3. Information containing beginning share price is given (by using Rp 610, - as the initial stock price).
2. Reading company background.
1. Participants are asked to respond the experiment psychological question to measure investor overconfidence characteristics.
7. Debriefing session Meanwhile, the experimental procedures for the end of sequence (EoS) disclosure pattern are executed as below:
6. Participants are asked to respond the manipulation check question, question to measure the participant basic understanding on financial statement analysis, investment demographic items.
5. Revaluing company stock price that will be bought as much as 18 times judgments for each information (accounting information and non- accounting information given).
16. Earnings Per Share of the company decreases than the last period.
17. Total Assets of the company decreases than the last period.
Eid Fitri in June 2015.
Research Participant
18. The company is facing problems from the employee related to the fulfillment of employee benefits in December 2015.
17. The company is facing demands of local communities related to the waste of the company that disrupt public interest in December 2015.
16. The company is facing problems related to the local communities demands related to the issue that there is inappropriate products sold to public in December 2015.
15. The company is facing demands of the local community related to the construction of factories in some regions in December 2015
14. There are internal parties of the company that misappropriate donations for education in December 2015.
13. The company only reports management interest and social services just as a form of accountability reports in December 2015.
12. The company has not used the code of conduct as the basis for the imposition of reward and punishment for employees in December 2015.
11. The company is facing the demands of the community because the use of the company's land disturbing public interest in December 2015.
10. The company is facing demands from other parties related to the infrastructure development for schooling assistance which use the land still in dispute in December 2015.
9. The company provides free trip as an appreciation for the community and to assist the government program in providing public transportation for
18. Net Profit Margin of the company decreases than the last period.
8. The company holds a social service in the form of free cataract surgery in order to participate in the movement of blind prevention of cataracts in June 2015.
7. The company is active in developing environmental programs to utilize and maximize the idle land by giving advisory services ranging from the planting, maintenance, harvesting, post-harvest processing to the raw material based on the factory standards and organic fertilizer production assistance in June 2015.
6. Company provides some donations to the community in the form of disaster relief, public infrastructures, worship places assistance and scholarships are given to students in June 2015.
5. Company provides free eyes health check-up and gives free glasses for students in June 2015.
4. Company provides health services to the people who suffer leprosy disease in June 2015.
3. Company gives donations for orphans from orphanages in some areas as well as daily packages for the poor people in June 2015.
2. Company provides services to the public in the form of clean water in the dry season in some regions in June 2015.
where the company provides training and mentoring in June 2015.
Persatuan Tunanetra Indonesia (Pertuni) to empower members of Pertuni in the cultivation of plants
1. The company cooperates with
Non-Accounting Information Items:
To test the hypothesis above, participants used in this study are undergraduate business students (management undergraduate or accounting undergraduate) who are studying and/or has completed investment and portfolio management and/or financial statement analysis courses. The undergraduate business students are chosen as the participants is based on the research conducted by Elliot et al. (2007) who explained that final year students had already have similar consideration pattern and similar investment decision making with non-professional investors in high and low task complexity.
Research Variables
Based on the related literature, hypothesis, and theoretical framework, the variable identification is as follows:
1. Dependent variable (X) is investment decision making. disclosure pattern (step by step and end of sequence), information order (++-- and --++), type of information (accounting information and non- accounting information), and investors overconfidence level.
Data analysis technique used in this research is normality test. In this case, normality test is utilized to ensure that the data used for the analysis is normally distributed. If the data is normally distributed, then the hypothesis testing will be continued by using parametric sample t- test. If the data is abnormally distributed, then the hypothesis testing will be continued by using non-parametric Mann- Whitney U test.
Criteria used for parametric sample t-test is: if the significant probability <0,05, it means that the null hypothesis is rejected, thus there is a different judgment. If the significant probability ≥0,05, it means that the null hypothesis is accepted, thus there is no different judgment. While in Mann-Whitney U test, if the significant probability ≥0,05, it means that the null hypothesis is accepted, thus there is no different judgment. If the significant probability <0,05, it means that the null hypothesis is rejected, thus there is a different judgment (Imam,2011:64).
HYPOTHESES TESTING RESULT AND DISCUSSION Demographic Data and Manipulation Check
Criteria for the subjects observed in this study are having knowledge in the field of investment and capital market and/or financial statement analysis, with the minimum cumulative GPA of 3.25. Based on criteria of the subject, the subjects in this study include accounting undergraduate students and management undergraduate students who are studying and/or has completed investment and capital market courses and/or financial statement analysis courses.
There were 120 subjects willing to participate in the study consisting of 116 accounting undergraduate students and 4 management undergraduate students. Eight subjects were not coming in the experimental instrument assignment. Therefore, 116 participants were assigned to the experimental scenarios until predetermined deadline. The participants were said to pass both of the manipulation check if they could meet the criteria determined by the researcher before. Manipulation check is intended to find out that the experimental task assigned was already understood and responded well and correctly by participants according to the guidelines presented by the experimenter. The predetermined criteria for the research participants which can be said to pass the calibration test, manipulation check and can be further analyzed were:
1. The participant’s correct answer related to the calibration test should be lower than their level of confidence.
2. The participant’s correct answer related to the manipulation check question should be minimum 2 out of 3 given questions.
3. The participa nt’s correct answer related to the question to measure the participant basic understanding on financial statement analysis and investment management should be minimum 2 out of 5 given questions.
4. The participants have fully respond and completed all of the experimental task assigned to them. If one of the above criterias were not met by the participants, they were said to fail in the calibration test and manipulation check and cannot be further analyzed. As a result, there were 21 subjects who did not pass the calibration test criteria and manipulation check, and hence they were not eligible for further analysis. Total subjects for further analysis and passed the manipulation check were
91 individuals consisting of 87 accounting undergraduate students and 4 management undergraduate students. The experimental research was held on December 2, 2016 in four different rooms (IIB402-IIB405). The execution of the research was started at 13.00 pm.
Examining The Effect of Information Order Presentation and Step-by-Step Information Disclosure Pattern on Investment Decision Making (Step-by- Step)
This research examines the effect of information order presentation and step- by-step information disclosure pattern on accounting information type and non- accounting information type by using long series information on investment decision making. The hypothesis 1 testing result is presented below:
Table 2: Independent Sample T-test Result for Hypothesis 1
Disclosure Pattern Information Order Mean Sig. 2-tailedStep-by-Step ++-- 568.33 0.014
Step-by-Step --++ 800 Source: SPSS version 22.0
Table 2 presented above shows the hypothesis testing result for the step-by step information disclosure pattern by using accounting information type for all participants. Based on the independent sample t-test hypothesis testing result above, it can be concluded that the significant probability value was 0.014, thus the hypothesis 1 is supported (0.014 < 0.05). It means that there is a different judgment between the participant who receives good news followed by bad news (++--) and the one who receives bad news followed by good news (--++) information order in the step-by-step disclosure pattern on accounting information type by using long series information on investment decision making. From the mean of the final judgment in the good news followed by bad news (++--) information order was lower than the final judgment in the bad news followed by good news (--++) information order, which were 568.33 for the order ++-- and 800 for the order of --
- . There is a significant difference between the participant who receives scenario I and the one who receives scenario II. It indicated that recency effect took place. This result is supported by figure 2 which shows the fishtail pattern in belief revision of the step by step information disclosure pattern by using accounting information type on long series information.
Table 3 Independent Sample T-test Result for Hypothesis 2 Disc. Pattern Order Mean Sig. 2- tailed
SbS ++-- 433.08 0.000 SbS --++ 870 Source: SPSS version 22.0 Table 3 presented above shows the hypothesis testing result for the step-by step information disclosure pattern by using non-accounting information type for all participants. Based on the independent sample t-test hypothesis testing result above, it can be concluded that the significant probability value was 0.000, thus the hypothesis 2 is supported (0.000 < 0.05). It means that there is a different judgment between the participant who receives good news followed by bad news (++--) and the one who receives bad news followed by good news (--++) information order in the step-by-step disclosure pattern on non-accounting information type by using long series information on investment decision making. From the mean of the final judgment in the good news followed by bad news (++--) information order was lower than the final judgment in the bad news followed by good news (--++) information order, which were 433.08 for the order of ++-- and 870 for the order of --++. There is a significant difference between the participant who receives scenario III and the one who receives scenario IV. It indicated that
recency effect took place. This result is
supported by figure 3 which shows the fishtail pattern in belief revision of the step by step information disclosure pattern by using non-accounting information type on long series information.
Discussion on The Effect of Information Order Toward The Final Judgment of Participants
Hypothesis 1 examines whether there is a different judgment between the participant who receives good news followed by bad news (++--) and the one who receives bad news followed by good news (--++) information order in the step- by-step disclosure pattern on accounting information type on investment decision making. This research is aimed to examine the existence of order effect on the step- by-step information disclosure pattern by using long series information. Table 4 shows the hypothesis 1 testing result.
Table 4: Hypothesis 1 Testing Result
Information Type Gained EffectStep by Step Information Disclosure Pattern (Accounting Information)
Recency Effect Hypothesis 2 examines whether there is a different judgment between the participant who receives good news followed by bad news (++--) and the one who receives bad news followed by good news (--++) information order in the step- by-step disclosure pattern on non- accounting information type on investment decision making. This research is aimed to examine the existence of order effect on the step-by-step information disclosure pattern by using long series information. Table 5 shows the hypothesis 2 testing result.
Table 5: Hypothesis 2 Testing Result
Information Type Gained EffectStep by Step Information Disclosure Pattern (Non-Accounting Information)
Recency Effect Both of the results of the hypothesis 1 and hypothesis 2 are not in accordance with the referred theory used by the researcher, that is, Belief Adjustment Model Theory (Hogarth and Einhorn, 1992). Belief Adjustment Model proposed by Hogarth and Einhorn (1992) predicts that by using long series information, simple information and the step-by-step information disclosure pattern, the primacy effect exists. Primacy
effect exists when the first information in
the order has a large effect on the individual belief.
Both of the hypothesis 1 and the hypothesis 2 testing result by using independent sample t-test show that the participants in this experimental research do not experience primacy effect in such order effect. Otherwise, the recency effect took place. Recency effect exist when the last information is the one that gives the largest effect on the final judgment. These results are supported by figure 4.1 which shows the fishtail pattern in belief revision of the step by step information disclosure pattern by using accounting information type and figure 4.2 which shows the fishtail pattern in belief revision of the step by step information disclosure pattern by using accounting information type on long series information. The fishtail pattern is a pattern formed when the recency effect took place (Hogarth and Einhorn: 1992).
Figure 4:
Fishtail Pattern on Accounting Information SbS Disclosure Pattern
The hypothesis 1 conducted with accounting information items on SbS scenarios. The results of the research showed that the mean of the final judgment in the good news followed by bad news (++--) information order was lower than the final judgment in the bad news followed by good news (--++) information order, which were 568.33 for the order ++-- and 800 for the order of --
- .
Based on the independent sample t- test hypothesis testing result above, it can be concluded that the significant probability value was 0.014, thus the hypothesis 1 is supported (0.014 < 0.05). It means that there is a different judgment between the participant who receives good news followed by bad news (++--) and the one who receives bad news followed by good news (--++) information order in the step-by-step disclosure pattern on accounting information type by using long series information on investment decision making. There is a significant difference between the participant who receives scenario I and the one who receives scenario II. It indicated that recency effect took place. 110 210 310 410 510 610 710 810 910 1010 1110 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 ++-- --++
Figure 4.2:
The hypothesis 2 conducted with accounting information items on SbS scenarios. The results of the research showed that the mean of the final judgment in the good news followed by bad news (++--) information order was lower than the final judgment in the bad news followed by good news (--++) information order, which were 433.08 for the order ++-- and 870 for the order of --
- .
Based on the independent sample t- test hypothesis testing result above, it can be concluded that the significant probability value was 0.000, thus the hypothesis 2 is supported (0.000 < 0.05). It means that there is a different judgment between the participant who receives good news followed by bad news (++--) and the one who receives bad news followed by good news (--++) information order in the step-by-step disclosure pattern on non- accounting information type by using long series information on investment decision making. There is a significant difference between the participant who receives scenario III and the one who receives scenario IV. It indicated that recency effect took place.