Keywords

Accounting

Abstract

This dissertation consists of three studies exploring nonprofessional investors' decision making. Technological advancements witnessed by the capital markets in recent years have caused significant changes to the dissemination and use of information, particularly by nonprofessional investors. Among these developments is the growth of social media that allows anyone to post information upon which others may rely and the availability of DAs that assist decision makers in evaluating the quality of information reported by an organization. The purpose of this dissertation is to investigate the benefits of using DAs that are capable assessing the quality of information reported to capital market participants and to investigate the effect of information retrieved from social media on nonprofessional investors' decisions. Study 1 highlights concerns over the ease of spreading video disclosures via social media outlets. Recent evidence from practice and research suggests that the trend of issuing video disclosures is growing and that investors are exposed to the risk of including deceptive information contained in those videos in their decision making process. The theoretical model introduced in this study suggests that investors can use deception detection DAs to identify deceptive behavior in video disclosures, and that the use of such DAs affects their perceptions of disclosure credibility. This study posits that management's pre-existing reputation affects investors' perceptions of disclosure credibility, and that the negative output of a deception detection DA can dilute the effect of management's pre-existing reputation on investors' perceptions of disclosure credibility. Using data from 376 nonprofessional investors, the findings support the proposed theoretical model and suggest that deception detection dilutes the effect of management's pre-existing reputation on investors' perceptions of disclosure credibility. The effect of management's pre-existing reputation on investors' perceptions of disclosure credibility is significantly weaker when the output of deception detection DA detects deception than when it fails to detect deception. Supplemental analyses suggest that the effect of deception detection is not limited to investors' perceptions of disclosure credibility, but also affects investors' willingness to invest. Deception detection dilutes the effect of management's pre-existing reputation on willingness to invest as well. These findings suggest that investors can mitigate the risks associated with video disclosures and improve their decisions by using deception detection DAs. Study 2 highlights concerns over the spread of linguistic manipulations in corporate disclosures. Recent evidence from the accounting literature suggests that managers strategically use linguistic manipulations and that investors unintentionally include the effect of these linguistic manipulations in their decisions. This study builds on the existing literature on linguistic manipulations and argues that providing investors with a DA that is capable of detecting linguistic manipulations can assist them in making investment decisions. The theoretical model introduced Study 2 suggests that the detection of linguistic manipulations (the occurrence of an expectation violation) moderates the effect of managers' incentives on investors' willingness to invest through disclosure credibility such that the effect of managers' incentive on investors' willingness to invest is expected to be weaker when the DA detects linguistic manipulations than when the DA fails to detect linguistic manipulations. Using data from 472 nonprofessional investors, the findings do not support the proposed theoretical model and suggest the effect of management incentive on investors' willingness to invest through disclosure credibility is not moderated by the detection of linguistic manipulations. These findings show that detecting linguistic manipulation has the same effect on managers with incentive to manipulate the language used corporate reports as those with no incentive to manipulate the language used in corporate reports. Study 3 highlights concerns over social media outlets that have enabled investors to communicate between themselves at an unprecedented rate. This study highlights the risk of using information retrieved from social media outlets and argues that investors are exposed to the risk of including erroneous information in their information set. This study uses the "Social Identification of the De-individuation Effect" model (SIDE) to argue that visual anonymity has an effect on investors' willingness to invest through their perceptions of disclosure credibility and that this effect depends on whether investors* have low or high social identification with the group of forum users. Using data from 401 nonprofessional investors, the findings do not support the proposed theoretical model. Nevertheless, findings from this study suggest that investors' social identification has an effect on their perceptions of disclosure credibility, and that social identification and visual anonymity have a joint effect on investors' willingness to invest. More precisely, investors with low social identification are more influenced by forum comments when they read the forum comments via text than when they view the forum comments via video; and, investors with high social identification are more influenced by forum comments when they view the forum comments than when they read the forum comments. While findings from this study provide support for the moderating role of social identification advanced in SIDE, the moderating role of social identification is in the opposite direction. Thus, this study fails to provide support for SIDE.

Notes

If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu

Graduation Date

2015

Semester

Summer

Advisor

Arnold, Vicky

Degree

Doctor of Philosophy (Ph.D.)

College

College of Business Administration

Degree Program

Business Administration; Accounting

Format

application/pdf

Identifier

CFE0005894

URL

http://purl.fcla.edu/fcla/etd/CFE0005894

Language

English

Release Date

August 2016

Length of Campus-only Access

1 year

Access Status

Doctoral Dissertation (Open Access)

Subjects

Business Administration -- Dissertations, Academic; Dissertations, Academic -- Business Administration

Restricted to the UCF community until August 2016; it will then be open access.

Included in

Accounting Commons

Share

COinS