Faked Depression: Comparing Malingering via the Internet,Pen-and-Paper,and Telephone Administration Modes

Introduction

The Diagnostic and Statistical Manual of Mental Disorders ¹ defines malingering as behavior motivated by external incentives, such as trying to obtain medication or avoid criminal charges. With such important incentives present, undetected malingering may be of particular concern for clinicians trying to accurately diagnose or treat individuals.

Yet, despite the growing use of the Internet and telephone for virtual therapy ² and telecounseling, ³ research investigating whether malingering differs as a function of psychological test administration mode is scarce: the first study comparing faking using Internet and pen-and-paper administration modes was only recently published. ⁴ Grieve and de Groot ⁴ found that depression could be faked equally regardless of whether the test was administered via the Internet or using pen and paper. The current research extended this approach and investigated for the first time whether malingered depression differs as a function of Internet, pen-and-paper, or telephone administration.

In line with previous research regarding respondents' ability to fake bad (e.g., Grieve and Mahar ⁵ ) and fake good (e.g., Grieve and Hayes ⁶ ), it was hypothesized that participants would be able to malinger as if experiencing depression when instructed. In line with the research paradigm of Grieve and de Groot, ⁴ the effect of administration mode on faked scores was investigated by comparing faked scores on the depression measure. As the current research included telephone administration for the first time and was therefore exploratory in nature, no specific hypotheses regarding any effects of test administration were proposed. However, in order to closely examine effects of administration mode, an examination of effect sizes was undertaken; this approach also aligned with the methodology of Grieve and De Groot. ⁴

Subjects and Methods

Results and Discussion

The manipulation check indicated that every participant had followed the instructions to malinger. Examples of responses included “Pretended I was really depressed” and “I answered like I was really low.” Thus, all cases were retained for analysis.

Descriptive statistics for the original and faked EDS across administration modes are presented in Table 1. The mean original scores across administration modes were similar to those reported in previous research. ^8,9 The mean faked scores across all three groups were well above the cutoff of 13 used in the EDS to indicate probable depression. ¹⁰ Reliability (Cronbach's α) for the original and faked depression scores was similar to previous research. ⁸

Analysis via within-groups t test indicated that as hypothesized, participants were able to successfully raise their scores significantly after instructed to fake as though suffering severe depression: t ₉₀=−29.74, p<0.001, r=0.90. This represented a large effect, ¹¹ with 81% of variance in faked depression scores explained by faking instruction.

To examine the effect of administration mode on faked depression, a between-groups analysis of variance was conducted. There was a significant difference in depression scores across the three administration modes: F _2,88=3.70, p=0.03. Post hoc analysis revealed that participants reported significantly lower faked depression scores over the telephone than in pen-and-paper format: t ₅₈=−2.59, p=0.01, r=0.32. The administration mode accounted for 10.24% of variance in the scores. Faked depression scores in the online group did not differ significantly from the telephone and pen-and-paper groups.

The current research examined only one measure of depression; therefore these results should not be overgeneralized. Also, it should be noted that the EDS does not contain questions on sensitive issues, which may differ in responsiveness as a function of administration mode. ^12,13 Although Knapp and Kirk ¹⁴ found no differences in responses to highly sensitive questions delivered via Internet, (touch-tone) telephone, and pen-and-paper, question sensitivity should be considered within a malingering context in future research. This point notwithstanding, the current findings provide promising initial evidence regarding the effect of administration mode on malingered depression and thus have implications for the telecounseling or virtual therapy domains. Results from the current study indicated that although there were statistically significant differences in malingered scores between telephone and pen-and-paper administrations, scores in all three administration modes met the cutoff for a provisional diagnosis of depression. ¹⁰ Hence, if scores from the online, telephone, or pen-and-paper group emerged in a real-life clinical setting, it is unlikely that they would lead to differing diagnoses or action by the clinician. Thus it seems that there may be no meaningful difference in malingering across administration modes: if an individual is malingering, he or she is able to do so regardless of testing modality.

These findings can be considered within the domain of online/telephone technologies in treatment. Previous research has indicated that online treatment is effective for depression. ¹⁵ In particular, for those in rural areas with poor access to health services, the use of telephone/online treatment can be useful. ¹⁶ For those counseling individuals via distance methods, results from the current study may appease some concern about the method of delivery, as individuals who intend to malinger can do so no matter the delivery mode. Given the benefits of online and telephone delivery for people in rural areas, ¹⁶ increased confidence in these administration modes may provide real benefit for those who live in these areas and rely on distance services.