Purpose of the Study
The study’s purpose is to evaluate the prevalence of electronic cigarette (e-cigarette) use for vaporizing substances other than nicotine among college students. Using the terminology, it seeks “to assess the prevalence of e-cigarette use, reasons for use, perceived harm, and prevalence and predictors of OSUE” (Kenne et al., 2017, p. 1). The research question: “What is the frequency of e-cigarette use for substances other than nicotine among college students, what are their reasons for this usage, and how do they perceive the harm associated with it?”. The hypothesis prediction: “There is a correlation between current tobacco use and an increased likelihood of using other substances in e-cigarettes (OSUE) among college students.”
Research Methods
Experimental or Observational
The study employs an observational design – researchers did not manipulate variables but rather surveyed participants to collect data. An example of the observational nature is the analysis of self-reported reasons for using e-cigarettes without intervention from the study conductors (Kenne et al., 2017). For instance, the authors state that “respondents were asked questions that included demographic characteristics, tobacco use (cigarette and smokeless), e-cigarette use and perceived harm of use, reasons for e-cigarette use” (Kenne et al., 2017, p. 2). The latter exemplifies that the researchers purely observed the behaviors and studied them without intervening or manipulating variables.
Methods
The method of data collection involved administering online (surveys) to undergraduate e-cigarette users from a large Midwestern university. The choice of an online survey is appropriate because it corresponds with the research question – it aims for a broad capture of self-reported behaviors and perceptions regarding e-cigarette use. The main reason is that it aligns well with the research question that seeks to understand self-reported behaviors and perceptions (Kenne et al., 2017). An online survey allows for the gathering of subjective data directly from the participants, which is essential in assessing their personal experiences and views regarding e-cigarette use.
Quantitative or Categorical
It should be noted that data collected in this study can be classified as both quantitative and categorical. The former includes any numerical data collected, such as the age of the respondents (mean age of 21.1 years), the perceived harm of e-cigarette use (rated on a scale), and past 30-day drug use (frequency of use) (Kenne et al., 2017). The authors also collected non-numerical data, such as race/ethnicity (white, black, other), gender (male or female), as well as tobacco cigarette smoking status (never smoker, former smoker, current smoker).
Weakness of the Data Collection Methods
However, a potential weakness in the data collection method includes the reliance on self-reporting, which can introduce bias. For example, participants can underreport or over-report their use of substances due to stigma or recall issues (Kenne et al., 2017). In addition, the sample comes from a single university, which essentially limits the generalizability of the findings. Considering them both, the collection makes the entire research non-generalizable and biased to a certain extent.
Data Analysis Methods
It should be stated that the study employed a binomial logistic regression data analysis method to investigate electronic cigarette (e-cigarette) usage among undergraduate students. The given method suits the study question intended to clarify the prevalence and predictors of using e-cigarettes for other substances (Kenne et al., 2017). The explanation is that the given regression model “predicts the probability that an observation falls into one of two categories of a dichotomous dependent variable based on one or more independent variables” (Laerd Statistics, 2018, para. 1). In the context of the study, such a method is effective for assessing the prevalence of behaviors and attitudes related to e-cigarette use, including OSUE.
Potential Weaknesses of the Data Analysis Methods
A specific statistical test utilized in the study is binomial logistic regression. For example, the authors state that “binomial logistic regression found that women were less likely to report OSUE;” hence, it only finds a correlation, but not causation (Kenne et al., 2017, p. 1). In other terms, it cannot determine if e-cigarette usage leads to OSUE or vice versa. In addition, the model assumes that the dependent variable is measured dichotomously (Laerd Statistics, 2018). The latter might not always be the case, but it works in this study because of how the authors framed the variables.
Key Demographics
Demographic data revealed that the study sample included 1542 undergraduate e-cigarette users from a Midwestern university. 35299 students were sent the online survey, and only 27% responded; among them, 5429 were undergrads, from which 1542 were selected as a sample (Kenne et al., 2017). The mean age was 21.1 years, with 55.3% women and 88.6% identifying as white (Kenne et al., 2017). Participants mostly grew up in suburban areas, and the inclusion criteria required respondents to be undergraduate students and e-cigarette users. Only undergraduates and “who reported lifetime use of e-cigarettes” were included (Kenne et al., 2017, p. 2). As a result, exclusion criteria were that non-undergraduates and those who did not report a lifetime of electronic cigarettes were excluded.
Key Findings
Results of the Research
The study found that a minority of the sample engaged in OSUE, with a significant number reporting the inhalation of cannabis. The data means that women tend to report lower OSUE, and individuals citing the use of e-cigarettes as being ‘cool or trendy’ are more likely to engage in OSUE (Kenne et al., 2017). The research findings indicated that 6.94% of the respondents used e-cigarettes for OSUE. Also, “current tobacco cigarette smokers were significantly more likely to report OSUE (51.0%) as compared with never (33.7%) and former (15.4%) smokers” (Kenne et al., 2017, p. 1).
Statistical significance was determined through binomial logistic regression, and such a method allows for the examination of relationships between categorical dependent and independent variables. The gender p-value is 0.004, and the OSUE-smoking status association is 0.014. Both p-values are lower than 0.05; therefore, it means that the differences are significant statistically. For other variables, such as race (p = 0.184), the p-value is bigger than 0.05; hence, it is not statistically significant.
Differences in the Results
There were marked differences in OSUE rates based on gender, smoking status, and reasons for e-cigarette use. For instance, women were less likely to report OSUE, whereas those using e-cigarettes for ‘cool or trendy’ reasons had higher OSUE rates (Kenne et al., 2017). The provided findings showcase how complex the behaviors of e-cigarette usage are when it comes to college students. No OSUE group (n = 1236) is larger than OSUE (n = 107). For the size of the difference, the OSUE rate among women was statistically lower than in men – women accounted for 43.0% of OSUE users compared to 57.0% in men, which showcases a significant gender-based difference in OSUE rates.
Limitations of the Study
The study’s limitation in population and sample size lies in its focus on a specific demographic. For example, it purely includes undergraduate students from a single Midwestern university (Kenne et al., 2017). In essence, it limits the generalizability of its findings to wider populations or other geographical areas. The use of binomial logistic regression allows for identifying associations between variables, such as gender and smoking status, with OSUE; however, the method cannot establish causality. For instance, it states: “binomial logistic regression found that women were less likely to report OSUE,” which is only a correlation (Kenne et al., 2017, p. 1).
The strength is that it is good for finding statistically significant associations between dichotomous factors, such as OSUE or non-OSUE. The cross-sectional design fails to track changes over time – it is a mere snapshot. For example, the findings are from 2014; hence, they do not account for what is happening today or after this year. This limitation exists because the research does not do the follow-up survey due to its design and the challenges of doing so with such a large body of respondents.
Major Conclusions from the Study
Compared to a study referenced in the article – Morean et al. (2015), which found 5.4% of high school students vaporizing cannabis using e-cigarettes – Kenne et al. (2017) showed a slightly higher OSUE rate among college students (6.94%). The given number difference likely means that there is an increase in OSUE with age. This study contributes to the scientific literature by showcasing the prevalence and predictors of OSUE among college students. Future investigations could include longitudinal studies to track OSUE behavior changes over time among this demographic.
References
Kenne, D., Fischbein, R. L., Tan, A. S. L., & Banks, M. (2017). The use of substances other than nicotine in electronic cigarettes among college students. Substance Abuse: Research and Treatment, 11, 1-8. Web.
Laerd Statistics. (2018). Binomial logistic regression using SPSS statistics. Web.
Morean, M. E., Kong, G., Camenga, D. R., Cavallo, D. A., & Krishnan-Sarin, S. (2015). High school students’ use of electronic cigarettes to vaporize cannabis. Pediatrics, 136(4), 611-616. Web.