UNIMAS PhD Viva Flashcards - Chapters 1 to 6

Comprehensive flashcards covering Chapters 1 through 6 for the UNIMAS PhD Viva on group motivational interviewing for problematic social media use.

What is social media addiction according to the viva notes?

A pattern of preoccupation and compulsive use causing impairment in social, academic, personal, work or well-being.

How is high social media use differentiated from social media addiction?

High use is time/frequency; addiction involves impaired control, preoccupation and negative consequences.

What is Motivational Interviewing (MI)?

A collaborative counselling approach that strengthens motivation for change by exploring and resolving ambivalence.

What is Group Motivational Interviewing (GMI)?

MI delivered in a group, using MI spirit and techniques to evoke motivation while using supportive group processes.

What is the Transtheoretical Model (TTM)?

A behaviour-change model describing movement through stages such as precontemplation, contemplation, preparation, action and maintenance.

Why was Motivational Interviewing (MI) chosen for social media use?

Social media has benefits and harms, so users may be ambivalent; MI is designed to work with ambivalence and strengthen self-directed change.

What was the main research gap identified in Chapter 1?

Malaysian medical-student studies were mainly cross-sectional, with little intervention evidence on GMI for problematic social media use.

Why study medical students specifically in this research?

Their academic, professional and well-being context makes digital behaviour relevant, while the cohort provides a defined educational setting.

What is the novelty of this study?

It prospectively examined GMI, motivation, readiness and app-recorded use in medical students rather than only cross-sectional associations.

Why is this research classified as a Medical Education PhD topic?

It examines learner behaviour, well-being and the educational environment using a behavioural intervention relevant to medical students.

Does the Bergen Social Media Addiction Scale (BSMAS) diagnose a clinical disorder?

No. It is a screening scale for addiction-related symptoms, not a clinical diagnostic interview.

How should a candidate defend the thesis title phrasing 'with social media addiction'?

Say it refers to the phenomenon studied, but 'social media addiction symptoms/problematic use' is more precise because not all screened above threshold.

Why discuss smartphone addiction when the focus of the thesis is social media?

Smartphones are a major access route and provide broader digital-addiction context, but smartphone and social media addiction are not identical.

What are Griffiths' six addiction components?

Salience, mood modification, tolerance, withdrawal, conflict and relapse.

Is social media addiction recognized as a formal DSM-5-TR diagnosis?

No. It is a research construct; problematic social media use is often a more cautious term.

Why can social media addiction relate to depression or anxiety?

The relationship may be bidirectional: distress may increase use, while problematic use may worsen distress.

Why are social media addiction prevalence estimates so heterogeneous in literature?

Different populations, instruments, cut-offs, platforms, sampling methods, cultures and time periods.

Why did the study measure both BSMAS and app-recorded screen time?

They capture different aspects: addiction symptoms versus behavioural duration, allowing triangulation.

Why should only a partial correlation be expected between BSMAS and screen time?

Time spent and addiction symptoms overlap but are not equivalent; heavy use can be purposeful and lower use can still be problematic.

Why choose Motivational Interviewing (MI) rather than Cognitive Behavioural Therapy (CBT)?

MI was chosen for ambivalence and motivation; the study did not claim MI is superior to CBT.

What are the four components that constitute the 'MI spirit'?

Partnership, acceptance, compassion and evocation, with respect for autonomy.

In Motivational Interviewing (MI), what does the acronym OARS stand for?

Open questions, affirmations, reflective listening and summaries.

How is ambivalence defined in Motivational Interviewing (MI)?

Simultaneously having reasons to change and reasons to continue the current behaviour.

What is the difference between change talk and sustain talk in MI?

Change talk favours change; sustain talk favours maintaining the current behaviour.

Are Motivational Interviewing (MI) and the Transtheoretical Model (TTM) the same theory?

No. They are distinct but conceptually compatible; MI can support movement in readiness to change.

Why did the study focus mainly on stages of change rather than the full TTM?

Stages offered a practical way to assess readiness, but the study was not a full test of the entire TTM.

What relationships did the study's conceptual framework propose?

GMI as the intervention, with BSMAS, motivation, readiness and app-recorded use as related measured outcomes; mediation was not formally tested.

Why was the study design quasi-experimental rather than fully experimental?

Because the final allocation of intact posting groups to conditions was non-random.

Why were intact posting groups used for allocation?

To reduce contamination between students and fit the existing academic timetable.

In this research design, what were the unit of assignment and unit of analysis?

The unit of assignment was the academic posting group; the unit of analysis was the individual student.

Why is the unit-of-assignment and unit-of-analysis mismatch important?

Students within a posting group may be correlated, so individual-level analysis can overstate precision if clustering is ignored.

What is clustering in group-based intervention studies?

Participants within the same group may be more similar to one another than participants from different groups.

What statistical consequences arise if clustering is ignored?

Standard errors may be underestimated, confidence intervals too narrow and Type I error risk increased.

What is a multilevel mixed model?

A model with fixed and random effects that can account for repeated measures within students and students within clusters.

What would the hierarchical levels be in a multilevel model for this study?

Level 1: repeated time points; Level 2: students; Level 3: posting groups.

Why did the analysis not implement a posting-group random effect?

Only four clusters were available, making cluster-level variance estimation unstable and unreliable.

Why were students not individually randomised to conditions?

It could increase contamination because students within the same posting group interact closely.

What design would be a stronger alternative for future research?

A larger cluster-randomised trial with many posting groups and appropriate cluster-adjusted or multilevel analysis.

How was the sample size calculated for the study?

G*Power, $\alpha = .05$, power $= .95$, effect size $d = 1.08$, giving $24$ per group; inflated for $30\%$ attrition to about $34$ per group.

What does an alpha level of $\alpha = .05$ represent?

A $5\%$ Type I error rate under the null hypothesis.

What does a planned power of $.95$ indicate?

If the assumed effect is real and assumptions hold, about a $95\%$ chance of detecting it statistically.

What is the meaning of 'nominal 95% power'?

Theoretical power based on the planned assumptions; actual power may be lower if the true effect is smaller.

Why was an assumed effect size of $d = 1.08$ potentially optimistic?

It came from a related but different intervention, population, target behaviour and comparator.

What are the methodological purpose and trade-offs of using an active psychoeducational control?

It controls for attention and information, but it may itself improve awareness and reduce between-group differences.

What is intervention fidelity, and why is its omission critical in this study?

Fidelity is the extent to which the intervention was delivered as intended; without independent fidelity checks, a null effect could reflect either an ineffective intervention or imperfect delivery.

What was the key finding for the primary outcome (BSMAS)?

No significant group $\times$ time interaction; GMI did not demonstrate superiority over control on the primary outcome.

What does a non-significant group $\times$ time interaction statistically mean?

The pattern of change over time did not differ statistically between GMI and control.

Does a statistically significant main effect of time mean that GMI worked?

No. It means scores changed over time across both groups, not that GMI caused the change.

What is the difference between the main effect of group and the main effect of time?

Main effect of group is the overall difference between GMI and control averaged across time points; main effect of time is the overall change across $T1$, $T2$ and $T3$ averaged across groups.

What does partial eta squared measure in this analysis?

The magnitude of an effect relative to that effect plus its associated error variance.

Can a finding of $p > .05$ prove that no intervention effect exists?

No. It means insufficient evidence against the null; effect size and confidence intervals also matter.

What was the baseline prevalence of social media addiction symptoms in the study cohort?

$49$ of $126$ participants, or $38.9\%$, screened above the BSMAS threshold.

What did the exploratory subgroup analysis of baseline BSMAS severity reveal?

Low-baseline students changed little; high-baseline students showed larger numerical reductions, but no significant group $\times$ time interaction.

Why is a floor effect a plausible explanation for low-baseline students showing little change?

They had less room for further reduction.

What were the findings regarding app-recorded screen time?

A significant group $\times$ time interaction showed a larger short-term reduction in GMI, not sustained at follow-up.

How can screen time show improvement while BSMAS scores do not?

Screen time measures duration; BSMAS measures addiction symptoms. They are related but distinct constructs.

What were the study findings regarding IAIMS motivation?

GMI showed a larger short-term increase, but the between-group difference was not sustained at follow-up.

What changes occurred regarding participants' stages of change?

GMI shifted participants toward later stages post-intervention, but this was not sustained at two months.

What was the correlation observed between BSMAS scores and app-recorded screen time?

Weak positive at baseline ($r_s \approx .26$), then very small and non-significant post-intervention and at follow-up.

Did BSMAS scores differ significantly between male and female participants?

No statistically significant difference; females were slightly higher, with a small effect.

Why should a researcher avoid focusing exclusively on significant secondary outcomes?

The primary outcome was null and secondary findings were multiple and exploratory, so overemphasis would overstate efficacy.

What is the central message of the discussion in Chapter 5?

GMI did not beat psychoeducation on BSMAS; short-term secondary changes occurred but were not sustained.

Why might BSMAS scores have fallen across both experimental and control groups?

Repeated assessment, self-monitoring, psychoeducation, regression to the mean, temporal influences or social desirability.

What is the distinction between measurement reactivity and the mere-measurement effect?

Measurement reactivity occurs when measurement itself changes awareness or behaviour; the mere-measurement effect means simply asking repeatedly about a behaviour may influence that behaviour even without the intended intervention.

How does regression to the mean explain score shifts across time?

Unusually high or low scores tend to move closer to the average on repeated measurement due to natural variability.

What are temporal influences in the context of this medical student cohort?

Time-related factors such as exams, holidays, workload or posting intensity that may affect outcomes independently of intervention.

Why does a null BSMAS interaction not imply that GMI is universally ineffective?

It means this study did not demonstrate superiority under this design, population and comparator.

What is an intermediate psychological construct, and how does it relate to motivation?

A psychological process that may change earlier in the pathway to behaviour change, but is not the final outcome itself. Motivation is an intermediate construct that may precede sustained symptom change.

Can increased motivation be claimed to cause the reduction in screen time?

No. Both changed, but mediation was not tested, so causation between them cannot be claimed.

Why might app-recorded screen time have rebounded at the two-month follow-up?

Structured support, goal setting and accountability ended, allowing previous habits to re-emerge.

What is a measurement limitation of evaluating change using categorical stages?

A participant can improve within a stage without crossing into another category.

What is the overall conclusion of the PhD thesis presented in Chapter 6?

GMI did not show sustained superiority on BSMAS; short-term motivational/behavioural signals were exploratory and not maintained.

What is the primary theoretical contribution of the research?

Motivation, readiness, behaviour and addiction symptoms may change at different rates rather than moving together.

How do the study findings connect to Self-Determination Theory (SDT)?

GMI is autonomy-supportive, but autonomy, competence and relatedness were not directly measured.

What is the practical recommendation regarding routine implementation of GMI?

GMI may help early motivation, but evidence is insufficient for routine stand-alone implementation; further stronger trials are needed before recommending routine use.

What rationale justifies recommending booster sessions in future interventions?

Short-term effects faded; boosters are a plausible way to support maintenance but were not tested.

Why is an open-label design considered a study limitation?

Knowledge of allocation can alter behaviour and reporting through expectancy, Hawthorne and social-desirability effects.

Why is delivering both arms using the same facilitator a limitation?

Facilitator style or MI-consistent communication could influence both conditions and cause contamination.

Why is the use of the IAIMS identified as a limitation?

It measures motivation for problematic internet use broadly, not social media specifically, reducing construct specificity.

What key changes should be implemented if redesigning this trial?

More clusters, cluster randomisation, multilevel analysis, longer follow-up, fidelity checks, better measures and prespecified missing-data handling.