November 4th UV on

n = 3

n = 3

November 12th UV on

November 18th UV off

n = 3

n = 3

NWCA bacteria breathing zone sample 3 = TNTC

November 25th UV Off

n = 3

n = 3

UV On & Off Comparison West Wing

Statistical Analysis

West Wing Vent Height - Bacteria

## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$concentration
## W = 0.65568, p-value = 2.372e-09
## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$log_concentration
## W = 0.98066, p-value = 0.6065

The log-transformed concentrations are approximately normal.

## 
##  Welch Two Sample t-test
## 
## data:  log_concentration by uv
## t = -1.9496, df = 44.978, p-value = 0.05747
## alternative hypothesis: true difference in means between group off and group on is not equal to 0
## 95 percent confidence interval:
##  -0.94106269  0.01531909
## sample estimates:
## mean in group off  mean in group on 
##          3.617556          4.080428

No statistically significant reduction in concentration was observed between UV off and on.

## # A tibble: 2 × 3
##   uv    mean_concentration median_concentration
##   <chr>              <dbl>                <dbl>
## 1 off                 48.5                   30
## 2 on                  85.4                   55
## [1] -75.96567

The bacterial concentration with UV on was ~76% higher than with UV off.

West Wing Vent Height - Fungi

## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$concentration
## W = 0.61215, p-value = 5.005e-10
## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$log_concentration
## W = 0.96963, p-value = 0.245

The log-transformed concentrations are approximately normal.

## 
##  Welch Two Sample t-test
## 
## data:  log_concentration by uv
## t = -3.5531, df = 45.994, p-value = 0.0008929
## alternative hypothesis: true difference in means between group off and group on is not equal to 0
## 95 percent confidence interval:
##  -1.4036904 -0.3884282
## sample estimates:
## mean in group off  mean in group on 
##          2.376771          3.272830

UV on concentrations were significantly higher than UV off

## # A tibble: 2 × 3
##   uv    mean_concentration median_concentration
##   <chr>              <dbl>                <dbl>
## 1 off                 14.6                   10
## 2 on                  39.0                   25
## [1] -167.1429

The fungal concentration with UV on was ~167% higher than with UV off.

West Wing Breathing Zone - Bacteria

## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$concentration
## W = 0.59535, p-value = 3.716e-10
## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$log_concentration
## W = 0.97143, p-value = 0.3001

The log-transformed concentrations are approximately normal.

## 
##  Welch Two Sample t-test
## 
## data:  log_concentration by uv
## t = -2.575, df = 44.707, p-value = 0.01341
## alternative hypothesis: true difference in means between group off and group on is not equal to 0
## 95 percent confidence interval:
##  -0.9848439 -0.1202842
## sample estimates:
## mean in group off  mean in group on 
##          3.623586          4.176151

Bacterial concentrations were significantly higher with UV on than UV off.

## # A tibble: 2 × 3
##   uv    mean_concentration median_concentration
##   <chr>              <dbl>                <dbl>
## 1 off                 45                     40
## 2 on                  89.4                   65
## [1] -98.61111

The bacterial concentration with UV on was ~99% higher than with UV off.

West Wing Breathing Zone - Fungi

## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$concentration
## W = 0.90471, p-value = 0.0008952
## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$log_concentration
## W = 0.94772, p-value = 0.03247
## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  concentration by uv
## W = 111.5, p-value = 0.0002323
## alternative hypothesis: true location shift is not equal to 0
## 
##  Welch Two Sample t-test
## 
## data:  log_concentration by uv
## t = -4.0882, df = 44.217, p-value = 0.0001807
## alternative hypothesis: true difference in means between group off and group on is not equal to 0
## 95 percent confidence interval:
##  -0.9621385 -0.3268140
## sample estimates:
## mean in group off  mean in group on 
##          2.581490          3.225966

Both parametric (Welch t-test on natural log-transformed data, p = 0.00018) and non-parametric (Wilcoxon rank-sum test, p = 0.00023) analyses confirmed a significant increase in bacterial concentrations with UV on.

## # A tibble: 2 × 3
##   uv    mean_concentration median_concentration
##   <chr>              <dbl>                <dbl>
## 1 off                 15.2                   15
## 2 on                  28.1                   25
## [1] -84.93151

The fungal concentration with UV on was ~85% higher than with UV off.

Nov 12 East & West Wing Vent Height - Bacteria

## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$concentration
## W = 0.92994, p-value = 0.2723
## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$log_concentration
## W = 0.86964, p-value = 0.03333

Normality assumptions not met.

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  concentration by uv
## W = 9, p-value = 0.0383
## alternative hypothesis: true location shift is not equal to 0

There was a statistically significant difference in bacterial concentration between UV on and UV off.

## # A tibble: 2 × 3
##   uv    mean_concentration median_concentration
##   <chr>              <dbl>                <dbl>
## 1 off                 32.5                   20
## 2 on                  78.9                   75
## [1] -142.735

The bacterial concentration with UV on was ~143% higher than with UV off.

Nov 12 East & West Wing Vent Height - Fungi

## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$concentration
## W = 0.75514, p-value = 0.001033
## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$log_concentration
## W = 0.98816, p-value = 0.9982

The log-transformed data are consistent with a normal distribution.

## 
##  Welch Two Sample t-test
## 
## data:  log_concentration by uv
## t = -0.49069, df = 12.989, p-value = 0.6318
## alternative hypothesis: true difference in means between group off and group on is not equal to 0
## 95 percent confidence interval:
##  -1.2690904  0.7993246
## sample estimates:
## mean in group off  mean in group on 
##          3.359825          3.594708
## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  concentration by uv
## W = 23.5, p-value = 0.723
## alternative hypothesis: true location shift is not equal to 0

There is no statistically significant difference between UV off and UV on.

## # A tibble: 2 × 3
##   uv    mean_concentration median_concentration
##   <chr>              <dbl>                <dbl>
## 1 off                 35.8                   25
## 2 on                  60                     40
## [1] -67.44186

The fungal concentration with UV on was ~67% higher than with UV off.

Nov 12 East & West Wing Breathing Zone - Bacteria

## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$concentration
## W = 0.95072, p-value = 0.5359
## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$log_concentration
## W = 0.87225, p-value = 0.03642

Normality assumptions not met.

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  concentration by uv
## W = 10, p-value = 0.05099
## alternative hypothesis: true location shift is not equal to 0

There is no statistically significant difference between UV off and UV on.

## # A tibble: 2 × 3
##   uv    mean_concentration median_concentration
##   <chr>              <dbl>                <dbl>
## 1 off                 34.2                 32.5
## 2 on                  67.2                 70
## [1] -96.74797

The bacterial concentration with UV on was ~97% higher than with UV off.

Nov 12 East & West Wing Breathing Zone - Fungi

## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$concentration
## W = 0.84697, p-value = 0.01574
## 
##  Shapiro-Wilk normality test
## 
## data:  filtered_data$log_concentration
## W = 0.95893, p-value = 0.6738
## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  concentration by uv
## W = 26, p-value = 0.952
## alternative hypothesis: true location shift is not equal to 0
## 
##  Welch Two Sample t-test
## 
## data:  log_concentration by uv
## t = 0.0066502, df = 6.4187, p-value = 0.9949
## alternative hypothesis: true difference in means between group off and group on is not equal to 0
## 95 percent confidence interval:
##  -0.9295146  0.9346614
## sample estimates:
## mean in group off  mean in group on 
##          3.474321          3.471748

There is no statistically significant difference between UV off and UV on.

## # A tibble: 2 × 3
##   uv    mean_concentration median_concentration
##   <chr>              <dbl>                <dbl>
## 1 off                 44.2                 27.5
## 2 on                  34.4                 35
## [1] 22.01258

The fungal concentration with UV on was ~22% lower than with UV off.