This report evaluates differences in time-of-completion of aseptic technique transfer in two different Microbiology Laboratory (MCB2010L) sessions (1419, 1441) taught during the Fall semester of the academic year 2016-2017 at Miami Dade College. The students, with the aid of a classmate, recorded themselves performing a slant-to-broth aseptic transfer. The recorded video of the aseptic transfer technique was then uploaded through a Dropbox link sent via email by the professor.
Note: The RMarkdown for this report can be found here.
At the time of creating this report,
## [1] "15 students submitted the video in MCB2010L section 1419"
## [1] "15 students submitted the video in MCB2010L section 1441"
Section 1419 has 20 students enrolled, while section 1441 has 22; therefore, section 1419 and 1441 had 5 and 7 students, respectively, that have not submitted their aseptic transfer video at the moment of elaborating this report.
What is the time-to-completion mean and 95% confidence intervals?
When considering the both MCB2010L lab sections,
## [1] "The mean time-to-completion was 84.2"
## [1] "The 95% confidence intervals were 72.6373345855642 95.7626654144358"
When considering both sections together, on average the students’ time-to-completion was 84 seconds, and 72 and 96 seconds as 95% confidence intervals from the mean. This suggests that most students, on average, took 1 minute and 20 seconds to complete the aseptic transfer, and this time-to-completion mean would range between 1 minute and 12 seconds and 1 minute and 30 seconds if calculaated from a random sample students that submitted the video.
To evaluate if there were differences in time-to-completion per MCB2010L sections, the means and median time-of-completion per lab section were calculated and compared.
| section | mean_time | median_time |
|---|---|---|
| 1419 | 73 | 76 |
| 1441 | 95 | 87 |
As seen in table 1 above, on average, students in section 1419 carried out the aseptic transfer in less time: section 1419’s mean was 24% and median time-of-completion 13% less, respectively, than section 1441. To further evaluate the time-of-completion by laboratory section, the statistical significance of the differences seen in table 1 were examined. First, we evaluated if the time-to-completion values for each section were parametrically distributed a determined by the Shapiro-Wilk normality test.
##
## Shapiro-Wilk normality test
##
## data: video_1419_1$time
## W = 1, p-value = 1
##
## Shapiro-Wilk normality test
##
## data: video_1441_1$time
## W = 0.9, p-value = 0.04
Although the time-of-completion for section 1419 was parametrically distributed, section 1441 was not. For this reason, the time-of-completion differences by lab section would be evaluated by the Wilcoxon rank sum test (not by the t-test).
##
## Wilcoxon rank sum test with continuity correction
##
## data: time by section
## W = 70, p-value = 0.06
## alternative hypothesis: true location shift is not equal to 0
As determine by the Wilcoxon rank sum test above, although the magnitude of the time-to-completion difference of section 1419 compared to section 1441 was 13%, the difference was borderline not statistically significance (p = 0.06).
To evaluate if being at a specific working station influenced the students’ time-of-completion for the aseptic transfer, the dataset was 1st stratified by working station, and then by both working station and table.
| table | mean_time | median_time |
|---|---|---|
| 1 | 94 | 87 |
| 2 | 85 | 82 |
| 3 | 73 | 69 |
| 4 | 72 | 60 |
| 5 | 95 | 80 |
When considering both sections together(table 2), on average, there appears that tables 3 and 4 completed the aseptic transfer in the shortest time, while 1 and 5 spent the most time completing the aseptic transfer.
To determine if these differences hold true after in both sections, the dataset was then adjusted for both working station and MCB2010L section.
pander(video %>% group_by(table, section) %>% summarize (mean_time = mean(time),
median_time = median(time)) %>%
arrange(mean_time, median_time),
caption = "Table 3: Time-of-completion by working station and MCB2010L section")
| table | section | mean_time | median_time |
|---|---|---|---|
| 2 | 1419 | 47 | 51 |
| 4 | 1441 | 55 | 55 |
| 3 | 1441 | 63 | 65 |
| 4 | 1419 | 76 | 66 |
| 5 | 1419 | 80 | 78 |
| 3 | 1419 | 83 | 80 |
| 1 | 1419 | 85 | 85 |
| 1 | 1441 | 96 | 95 |
| 2 | 1441 | 114 | 94 |
| 5 | 1441 | 114 | 107 |
When comparing both sections by working stations (Table 3), the working station-section with the shortest time-of-completion was working station 2 from section 1419; the longest time-of-completion belonged to working station 5 from secion 1441. Although table 1 from both sections were among the lowest times-of-completions, table 5 from section 1419 is among the shortest times-of-completion. In addition, only tables 3 and 4 from section 1441 are among the fastest in completing the aseptic transfer. Therefore, difference patterns in time-of-completion between working stations seen when clustering both section does not holds when adjusting by MCB2010L sections.
After evaluating working stations, and section, it was worthwhile evaluating if gender influenced time of completion. The distribution of gender in both sections (for students that had submitted the video) were the following: section 1419, 10 females and 5 males; for section 1441, 8 females and 7 males.
| 1 | 2 | 3 | 4 | 5 | |
|---|---|---|---|---|---|
| F | 0 | 3 | 3 | 1 | 3 |
| M | 1 | 0 | 0 | 3 | 1 |
| 1 | 2 | 3 | 4 | 5 | |
|---|---|---|---|---|---|
| F | 3 | 2 | 2 | 1 | 0 |
| M | 1 | 2 | 1 | 0 | 3 |
Among the submitted videos, in section 1419 there were no females in working station 1, and no males i working station 2 and 3. For section 1441, again among the submitted videos, there were no females in working station 5 and no males in working station 4.
| gender | mean_time | median_time |
|---|---|---|
| F | 78 | 77 |
| M | 94 | 90 |
##
## Wilcoxon rank sum test with continuity correction
##
## data: time by gender
## W = 60, p-value = 0.04
## alternative hypothesis: true location shift is not equal to 0
As shown in table 6 above, female completed the aseptic transfer in shortest time, and the differences is statistically significant (p = 0.04) was determined by the Wilcoxon rank sum test.
| gender | section | mean_time | median_time |
|---|---|---|---|
| F | 1419 | 68 | 72 |
| F | 1441 | 89 | 80 |
| M | 1419 | 84 | 85 |
| M | 1441 | 101 | 96 |
Furthermore, the gender difference holds true after adjusting for MCB2010L sections (table 7), with females having comparable mean and median’s both section.
Going one step further, the gender-section differences seen in table 7 were further evaluated by working station as shown in the bar graph below.
Earlier, we reported that working station 2 from section 1419 and working station 4 from section 1441 had the shortest time-of-completion. Among the submitted videos, both of these working statinos were females students only. It was also reported that working station 5 from section 1441 reported the longest time-of-completion, and this working station was male students only.
This report evaluated the time-of-completing of the aseptic transfer exercise between two MCB2010L sections. Section 1419 reported shortest time-of-completions, and working stations in which there were female-only students performed the aseptic transfer technique in shortest times. The findings of this report suggest that it could be worthwhile implementing interventions in working stations in which there are male-only students or in which there are males.