This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Viewing Movie Violence & Interest in Guns (NCT03220412), a randomized na study. The primary outcome is Trigger Pulls.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03220412 |
| Phase | NA |
| Condition | Psychology, Social, Adolescent Behavior |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Single |
| Enrollment | 104 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Ohio State University |
| Results posted | Yes |
2. Clinical Question
Population
Children participating in a study of exposure to gun violence in movies and interest in real guns. The registry record does not report the participants’ ages or other eligibility details. It gives enrollment as 104 and reports group sizes of 54 in the experimental condition and 50 in the control condition.
Intervention
Viewing a movie with guns. The record describes the experimental condition as showing the movie as filmed and distributed, without editing the relevant scenes. National Treasure and The Rocketeer are named as movie options.
Comparator
Viewing a movie without guns. The control-condition description says the movie was edited to remove guns from scenes. The brief summary frames the comparison as the same movie with or without guns.
Primary question
Whether children who viewed a movie containing guns handled a real gun longer and pulled the trigger more times than children who viewed a movie without guns. The posted primary result concerns trigger pulls, measured as a count per child 20 minutes after the intervention. A primary result for handling duration is not reported in the registry record.
The research question is about a short-term behavioral response following a particular viewing experience. It is not a question about longer-term interest, later behavior, or the effects of media exposure outside this study. The registry’s official title states a directional conclusion, but the result described for the primary endpoint is an adjusted median for each condition. A direct estimate of the between-condition difference, and a statistical test of that difference, are not reported in the registry record. The clinical question therefore needs to be distinguished from the specific numerical result that is posted.
3. Trial Design
The registry describes a randomized, single-masked, parallel-group study. Children were assigned to an experimental condition involving a movie with guns or a control condition involving a movie edited to remove guns. The study is listed as completed, with enrollment of 104. The posted group sizes are 54 for the experimental condition and 50 for the control condition. The registry lists the phase as not applicable.
The study record gives a start date of 2015-07-15 and a primary completion date of 2016-01-01. It names Ohio State University as the sponsor. The listed conditions are Psychology, Social and Adolescent Behavior. These administrative details describe the registered study; they do not supply additional information about how participants were recruited or how random assignment was carried out.
The intervention was behavioral rather than a drug or medical procedure. The experimental condition viewed a movie with guns, and the control condition viewed a version from which guns had been edited. National Treasure and The Rocketeer are named in the experimental-condition description. The record does not fully explain how movie selection was handled for each participant, and it does not identify who was masked. “Single” is the registry’s masking designation; the identity of the masked party is not reported in the registry record.
The primary outcome was assessed 20 minutes after the intervention. The posted primary-result description uses a reduced Generalized Estimating Equation model, including participant gender and condition but not other control variables. Details about the randomization procedure, concealment, setting, and analysis population are not reported in the registry record. Those omissions limit how fully the design and conduct can be reconstructed from the registry entry alone.
4. Participant Flow and Accounting
The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.
Overall Study
| Stage | Experimental Condition | Control Condition |
|---|---|---|
| STARTED | 54 | 50 |
| COMPLETED | 54 | 50 |
| NOT COMPLETED | 0 | 0 |
5. Baseline Characteristics
The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.
| Characteristic | Unit | Experimental Condition | Control Condition | Total |
|---|---|---|---|---|
| Participants analysed | 54 | 50 | 104 | |
| Age, Categorical: <=18 years | Participants | 54 | 50 | 104 |
| Age, Categorical: Between 18 and 65 years | Participants | 0 | 0 | 0 |
| Age, Categorical: >=65 years | Participants | 0 | 0 | 0 |
| Age, Continuous | years | 9.7 (SD 1.5) | 10.14 (SD 1.5) | 9.9 (SD 1.5) |
| Sex: Female, Male: Female | Participants | 16 | 26 | 42 |
| Sex: Female, Male: Male | Participants | 38 | 24 | 62 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 1 | 2 | 3 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 53 | 48 | 101 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Asian | Participants | 2 | 0 | 2 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 14 | 17 | 31 |
| Race (NIH/OMB): White | Participants | 33 | 29 | 62 |
| Race (NIH/OMB): More than one race | Participants | 5 | 4 | 9 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Region of Enrollment: United States | Participants | 54 | 50 | 104 |
6. Primary Endpoint
The registry lists the primary outcome as Trigger Pulls, measured over 20 minutes after intervention.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | 95% Confidence Interval |
| Unit | Number of trigger pulls |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is the number of trigger pulls per child, assessed 20 minutes after the intervention. It is a count of an observed behavior rather than a self-reported measure of interest. The registry describes the reported condition-specific values as adjusted medians from a reduced Generalized Estimating Equation model. That model included participant gender and condition, but no other control variables.
In general, a median is the central value when observations are ordered: it is less influenced by unusually large observations than a mean. A count outcome can include many small values and some larger ones, so a median can be a useful way to describe the typical observed count. The registry does not report the full distribution of trigger-pull counts, so the spread, frequency of zero counts, and presence of unusually high counts cannot be assessed from the posted primary result alone.
The endpoint operationalizes one immediate behavior relevant to the study’s question. It does not measure how long a child handled a gun, even though handling duration appears in the study objective. A result for that measure is not reported in the registry record. Nor does a trigger-pull count, by itself, measure a child’s general attitude toward guns or predict behavior outside the study setting. The short interval between viewing and outcome assessment defines the scope of the endpoint’s interpretation.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The experimental-condition adjusted median was 2.8 trigger pulls per child, with a 95% confidence interval from 0.2 to 2.8. The control-condition adjusted median was 0.01, with a 95% confidence interval from 0.01 to 0.2. These are the condition-specific estimates reported for the primary endpoint, measured 20 minutes after the intervention.
The numerical estimates describe different central values in the two conditions, but they are not themselves a direct estimate of the between-condition contrast. The registry record does not report a contrast estimate, its confidence interval, or a p-value for the primary outcome. It therefore does not provide the inferential result needed to characterize the precision of the difference between conditions. The confidence intervals attached to the separate medians should not be treated as a substitute for a confidence interval around that contrast.
In general, a confidence interval communicates uncertainty around an estimate under the statistical model and procedure used. It does not mean that every value inside it is equally plausible, nor does it establish that a difference between groups is statistically distinguishable. The posted intervals describe the reported adjusted medians; the record does not provide enough information to reconstruct the uncertainty of a direct comparison. No additional statistical test should be inferred from the fact that the two point estimates differ.
The adjustment described is limited: the reduced model included participant gender and condition, and excluded other control variables. The registry does not report which other candidate variables were considered, why this reduced model was selected, or whether any model alternatives were assessed. The displayed estimates should therefore be read as results from the stated model, not as results adjusted for every possible characteristic or source of variation.
The registry’s study objective includes both handling duration and trigger pulls, but the posted primary result concerns trigger pulls. The record lists two outcome measures, while another endpoint is identified as having values missing or not estimable for a two-arm comparison. No secondary results are posted. The primary result consequently addresses only the available reported endpoint and should not be generalized to unreported outcomes.
Overall, the posted data show the two adjusted median counts and their intervals. They do not, on their own, establish the size or statistical uncertainty of a between-condition effect. The official title expresses a conclusion, but interpretation of the numerical result must remain anchored to the estimates, model description, and missing inferential details in the registry record.
9. Statistical Methods Explained
The primary result is described as an adjusted median from a reduced Generalized Estimating Equation model. The registry states that the model included condition and participant gender, but not other control variables. The reported quantity is therefore not simply identified as an unadjusted raw median. The record does not provide the model formula, the estimating procedure, the variance estimator, or details of any working correlation structure.
What does a median describe?
In general, a median locates the middle of an ordered set of observations. For a count endpoint, it gives a central summary in the original count scale and is less affected than a mean by very large observations. A median does not describe the full distribution: it does not reveal how many observations were zero, how much counts varied, or whether there were high-count values. Those features can matter when evaluating behavioral count data, but the distribution is not reported in the registry record.
The values reported are adjusted medians for the two conditions. Adjustment means that the model incorporates the included variable or variables when producing its estimates; it does not mean that every potential source of imbalance has been addressed. Here, the record specifically names participant gender and condition and says that other control variables were not included in the reduced model. It does not describe any further adjustments or explain the model-selection process.
What does a Generalized Estimating Equation model do?
In general, Generalized Estimating Equations (GEE) are a modeling approach for estimating population-average relationships when observations may be correlated within groups or clusters. The approach uses a specified mean model and accounts for a chosen pattern of dependence in its estimating and uncertainty procedures. It can be used for different outcome types, including counts. The registry’s use of GEE is reported, but the record does not explain the dependence structure, identify any clustering unit, or give the model’s distributional and variance specifications. It is therefore not possible to determine those details from the entry.
GEE estimates are often interpreted as average relationships at the population level rather than as subject-specific effects. That is a general property of the approach, not a further result about this study. The particular interpretation depends on the model specification and the data structure, details that are not fully. The result should be described using the registry’s stated adjusted-median language rather than assuming a more detailed model interpretation than the entry supports.
How should the confidence intervals be read?
In general, a confidence interval reflects uncertainty in an estimate under the method’s assumptions. The registry reports 95% confidence intervals for each condition-specific adjusted median: 0.2 to 2.8 for the experimental condition and 0.01 to 0.2 for control. These intervals concern the reported condition-specific estimates. The record does not report a confidence interval for a direct between-condition contrast, so the intervals cannot be used as if they were that contrast interval.
The confidence intervals also do not show the individual observations or all sources of uncertainty. Their interpretation depends on the model and calculation procedure, whose full details are not reported in the registry record. They should be read as the posted uncertainty summaries, not as a guarantee about the range of values that would occur for individual children.
What do the planning calculations represent?
In general, sample-size calculations translate assumptions about a target effect and statistical procedure into a planning illustration. The registry’s stated assumptions describe a two-sided significance level of 0.05, equal allocation, and a normal approximation. The illustration is based on the observed effect, with scenarios for detecting a stated fraction of that effect. These are planning assumptions, not observed trial outcomes and not observed power. They do not establish that the original study was designed using those assumptions. The registry record does not report the trial’s own sample-size justification.
Planning calculations based on an observed effect can be sensitive to the effect estimate used: an estimate from a study may differ from the underlying effect because of sampling variation. In general, smaller target effects require more information to detect than larger ones, all else being equal. This is a statistical principle, not an additional numerical result for this trial. The registry’s record does not report an interim-analysis plan or other details that would change how the calculations should be interpreted.
The sample-size material should be read as a planning aid rather than as evidence about the strength of the trial’s observed result. It does not replace the missing between-condition estimate and inferential statistics for the primary endpoint. No planning value should be interpreted as the number of children actually observed in either condition; the posted group sizes are reported separately in the trial record.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists two outcome measures, but only the trigger-pull endpoint has the primary result. Another endpoint is flagged as having values missing or not estimable for a two-arm comparison. No secondary results are posted. The record does not explain whether the unavailable endpoint reflects missing observations, a modeling issue, a registry display limitation, or another reason.
Multiplicity concerns arise when several outcomes or statistical comparisons are tested, because repeated testing can increase the chance of a false-positive result. In general, investigators may prespecify a primary outcome, use an adjustment for multiple comparisons, or interpret secondary analyses as exploratory. The registry record does not report a multiplicity strategy, a hierarchy among outcomes beyond the designation of the primary endpoint, or a family of statistical tests. No adjustment or lack of adjustment should be assumed.
Missing-data methods are also not reported in the registry record. It does not state how missing outcomes were handled, whether all randomized participants contributed to the primary analysis, or whether any imputation or sensitivity analysis was used. Group sizes are reported, but the analysis population and endpoint-specific numbers contributing to the fitted model are not clearly reported. The posted group sizes should not be treated as confirmation that every participant had a complete outcome.
In general, the interpretation of a randomized comparison depends partly on whether outcome availability and analysis decisions preserve the intended comparison. Without a reported analysis population and missing-data approach, the extent to which missingness could influence the estimate cannot be assessed. These omissions do not alter the posted values; they limit the information available for evaluating their robustness.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Experimental Condition | 0 | 54 | 0.0% |
| Control Condition | 0 | 50 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Experimental Condition | 0 of 54 |
| Control Condition | 0 of 50 |
The registry reports zero serious adverse events and zero other adverse events in the experimental condition, among 54 at risk. It also reports zero serious adverse events and zero other adverse events in the control condition, among 50 at risk. No serious adverse-event terms are listed. Death counts are not reported in the registry record.
These figures describe what was recorded in the safety fields. They should not be read as proof that no adverse experiences occurred or that the intervention is risk-free. The record does not describe how adverse events were defined, collected, monitored, or attributed, and it does not provide narrative details about events or follow-up beyond the outcome timeframe. A zero count in a registry summary means that no events were entered in that category for the reported group; it does not establish that every possible harm was assessed or ruled out.
The two group-specific safety summaries are distinct from the primary behavioral endpoint. They do not provide evidence about the trigger-pull result’s statistical uncertainty, and they do not change the interpretation of its adjusted medians. The missing death counts should not be converted into a claim of zero deaths.
12. What the Result Does, and Does Not, Mean
The result supports a narrow description: the registry reports a higher adjusted median trigger-pull count in the experimental condition than in control, 20 minutes after the intervention. The values are 2.8 with a 95% confidence interval from 0.2 to 2.8 for the experimental condition, and 0.01 with a 95% confidence interval from 0.01 to 0.2 for control. The model included participant gender and condition, and excluded other control variables.
The registry does not report a direct between-condition estimate, confidence interval, or p-value. The difference between the two point estimates is not itself a complete statistical analysis of the comparison. The reported intervals concern condition-specific medians and do not supply the uncertainty of the contrast. Accordingly, the posted information does not establish whether the between-condition comparison meets any particular statistical criterion.
The study’s question concerns children’s short-term behavior after exposure to a movie with guns versus a movie edited to remove guns. The primary result is not a measure of long-term interest, general attitudes, or behavior in settings beyond the study. It also does not report the handling-duration outcome mentioned in the objective. The registry title states a broad directional claim, but the numerical result available for interpretation is limited to trigger pulls.
In general, random assignment can support a causal comparison when the assignment and analysis are carried out appropriately, but interpretation also depends on trial conduct, outcome measurement, missing data, and analysis decisions. The registry does not report enough detail to evaluate all of those elements. This does not negate the posted result; it defines what can and cannot be inferred from the record alone.
The result should therefore be treated as a reported, model-adjusted description of one short-term behavioral endpoint, rather than a complete account of the study’s findings or a conclusion about all effects of media exposure. The uncertainty and scope of any causal conclusion cannot be fully evaluated without the direct comparison statistics and further methodological details, which are not reported in the registry record.
13. Why This Trial Matters Statistically
This study illustrates why the form of a reported result matters. A study question usually concerns a comparison between conditions, while registries may report condition-specific estimates. Here, the primary outcome is reported as an adjusted median for each condition, with a confidence interval for each. Those estimates describe the groups, but they are not the same as a direct estimate of the difference between groups. Statistical interpretation is clearest when the contrast itself and its uncertainty are reported.
The endpoint is a behavioral count measured shortly after the viewing intervention. In general, count outcomes can have distributions that are asymmetric or concentrated near small values. A median can give a useful central summary, but additional distributional detail helps readers understand what that summary represents. The registry does not provide such detail, so the posted medians should not be taken as a complete description of individual responses.
The entry also demonstrates the importance of matching claims to the endpoint and time horizon actually reported. The study objective refers to both handling duration and trigger pulls, but the primary posted result concerns trigger pulls 20 minutes after the intervention. It cannot stand in for an unreported result on handling duration or for longer-term behavior.
Finally, methods and transparency affect how readers assess a result. The registry identifies a reduced GEE model and the covariates included, but does not report a contrast estimate, a test result, a missing-data approach, or a multiplicity plan. Sample-size illustrations are based on stated assumptions and the observed effect, rather than being evidence of observed power or the original design rationale. Together, these features make the record useful for describing what was posted while also showing why a result should be interpreted in proportion to the statistical detail available.
14. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The registry record does not report participants’ ages, eligibility criteria, recruitment methods, study setting, or the procedures used to generate and conceal random assignments. The masking designation is single, but the record does not identify who was masked. These omissions limit assessment of the study’s conduct and the population to which the findings might apply.
The movie descriptions are not fully detailed. The record names National Treasure and The Rocketeer and says that the control movie was edited to remove guns. It does not fully explain how movie options were selected or whether each participant’s viewing experience was otherwise identical. The brief summary describes the intended comparison as the same movie with or without guns, but further implementation details are not reported in the registry record.
The primary result gives adjusted medians and confidence intervals but no direct comparison estimate, p-value, or uncertainty interval for the contrast. The GEE model description is limited to its inclusion of participant gender and condition and exclusion of other control variables. The model specification, handling of any dependence, and rationale for the reduced model are not reported. The analysis population and approach to missing outcomes are also not reported.
The study objective includes handling duration, but a result for that measure is not reported. Another outcome is flagged as not available or not estimable for a two-arm comparison, and no secondary results are posted. The safety record reports zero serious and other adverse events in each group, but does not provide collection details or death counts. Finally, the registry’s planning assumptions are not evidence of the original sample-size rationale. These limitations constrain the level of detail with which the result can be evaluated.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Viewing Movie Violence & Interest in Guns — NCT03220412.
- Registry record last updated 2019-06-03; status last verified 2019-05.
- Results first posted 2019-05-10.
- Data retrieved from ClinicalTrials.gov on 2026-11-09.
- Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
The registry record describes a completed, randomized, single-masked, parallel-group behavioral study sponsored by Ohio State University. It lists enrollment of 104, with 54 participants in the experimental condition and 50 in control. The experimental condition involved viewing a movie with guns; the control condition involved a movie edited to remove guns. The record gives a start date of 2015-07-15 and a primary completion date of 2016-01-01.
The primary endpoint was the number of trigger pulls per child 20 minutes after the intervention. The posted adjusted medians were 2.8 for the experimental condition and 0.01 for control, with respective 95% confidence intervals of 0.2 to 2.8 and 0.01 to 0.2. The reduced GEE model included participant gender and condition.
A direct between-condition estimate and its inferential statistics are not reported in the registry record. Results for the handling-duration outcome are not reported, and no secondary results are posted. Safety fields record zero serious and other adverse events in both conditions; death counts are not reported.