This paper critically analyzes a randomized controlled trial by Kählke et al. (2019) examining the efficacy of an unguided internet-based self-help intervention for social anxiety disorder (SAD) in college students. The analysis evaluates the study's causal claims, experimental design, hypothesis, independent and dependent variables, participant characteristics, and treatment protocol. It also assesses the study's construct validity, external validity, temporal precedence, and efforts to control for third-variable problems. Supporting evidence from three additional studies is reviewed, collectively suggesting that internet and mobile-based interventions (IMIs) hold meaningful promise for addressing social anxiety, depression, and low mindfulness among college student populations.
The claim of the researchers is causal in nature, based on the fact that the study attempts to determine the efficacy of using internet- and mobile-based interventions (IMIs) to reduce social anxiety disorder through an experimental design (Kählke et al., 2019).
The design was experimental, given that a randomized control group and an experimental intervention group were used (Kählke et al., 2019).
The hypothesis was that in college students diagnosed with social anxiety disorder (SAD), using IMI would reduce symptoms as measured 10 weeks post-treatment on the Social Phobia Scale (SPS) and the Social Interaction Anxiety Scale (SIAS).
The primary conceptual independent variable — the one resulting from the researchers' manipulation — was social anxiety symptoms as operationalized by a pre-measured diagnosis. The primary conceptual dependent variable was the reduction of social anxiety symptoms, measured using independently validated scales.
All participants (n = 200) were college students over the age of 18 who met DSM-IV Axis I criteria for SAD. Exclusions were made for individuals with suicidal symptoms or a diagnosis of bipolar disorder, psychosis, or other severe mental disorders.
Participants in the experimental group received a cognitive behavioral intervention consisting of nine hourly text-based sessions involving exercises, diaries, and homework completed outside of class (Kählke et al., 2019). All students in the experimental group received the same intervention, while the control group was placed on a wait list for treatment.
The main results indicated that students in the experimental group self-reported a reduction in symptoms compared with those in the control group, as well as improvements on other self-reported assessment scales, including depression, anxiety, quality of life, and other secondary symptoms associated with social anxiety.
The hypothesis that using a purely internet-based intervention could reduce clinical social anxiety in college students — along with associated mental health issues such as depression — was supported.
Reliability was assessed using the Reliable Change Index of Jacobson and Truax and by measuring reduction in symptoms post-treatment.
Overall, the study demonstrates strong construct validity: all participants were assessed with independent, widely validated scales for social anxiety and met DSM diagnostic criteria, and participants were assigned to randomized control groups. The CBT intervention has been widely used to treat SAD previously and was specifically designed to address the disorder. Individuals likely to have other comorbidities were excluded, and the study attempted to reduce potentially confounding variables by focusing on a relatively narrow population from the same age group and social circumstances.
The study has moderately strong external validity. On one hand, it is likely that any intervention to help an individual with social anxiety is better than none, and CBT techniques have been widely and empirically validated outside of this study. On the other hand, the population was relatively homogeneous, which may not reflect the broader population even of college students.
Covariance is not likely a major concern in this study, given the high degree of randomization in assigning students to the experimental and control groups and the deliberate effort to reduce confounding variables related to mental health.
The study establishes temporal precedence given its lengthy duration (10 weeks), with pre-intervention and post-intervention testing of both groups.
The study also makes clear the researchers' efforts to address the third-variable problem by eliminating subjects with comorbidities for other psychological disorders as well as individuals taking mood-stabilizing medications.
"Study limitations and external validity improvements"
"Three corroborating IMI studies and conclusions"
Always verify citation format against your institution’s current style guide requirements.