Men Moving Forward Data Plotted

Summary of MMF Data Set

# Range of percentage difference in BIA versus DXA
range(COMB$DIFF_BONE_PCT)
## [1]  0 70
# Range of values from DXA subtracted from BIA
range(COMB$DIFF_BONE2)
## [1] -1.73  0.54
# Range of weights of the participants in kilograms
range(COMB$WEIGHT_DXA)
## [1]  56.4 171.0

Covid Survivor Study Data Plotted

Summary of CSS Data Set

# Range of percentage difference in BIA versus DXA
range(COMBCSS$DIFF_BONE_PCT)
## [1]  0 63
# Range of values from DXA subtracted from BIA
range(COMBCSS$DIFF_BONE2)
## [1] -1.67  0.38
# Range of weights of the participants in kilograms
range(COMBCSS$WEIGHT.x)
## [1]  44.4 152.6

##Discussion

Why is there such a substantial R-squared difference between the two?

As of now I’ve only explored the relationship between weight and percent fat as it pertains to the differnce in BMC in the Men Moving Forward Data Set.

Below I’ve run a multiple linear regression to determine if there’s a relationship between the BMC difference and percent fat differences in both the DXA and BIA measures

The histogram above shows a fairly normal distribution as well as the fitted versus residual dot plot
So ultimatley there’s not much of a relationship between the percent fat thats driving the difference in BMC differnce between the MMF and CSS data.
plot(fitted(model1), residuals(model1))+abline(h=0,lty=2)

## integer(0)
summary(model1)
## 
## Call:
## lm(formula = DIFF ~ PFAT_BIA + PFAT_DXA, data = COMBINED)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1.15808 -0.21918  0.02266  0.22688  1.01964 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)    
## (Intercept) -0.506851   0.153638  -3.299 0.001123 ** 
## PFAT_BIA     0.006073   0.006792   0.894 0.372150    
## PFAT_DXA     0.028393   0.008358   3.397 0.000801 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.386 on 232 degrees of freedom
## Multiple R-squared:  0.1582, Adjusted R-squared:  0.1509 
## F-statistic: 21.79 on 2 and 232 DF,  p-value: 2.124e-09

Here is the relationship between difference in BMC relative to DXA percent fat

Here is the relationship between difference in BMC relative to BIA percent fat

## 
## Call:
## lm(formula = y2 ~ x2, data = COMBINED)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1.07332 -0.25194  0.02295  0.22918  1.06913 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)    
## (Intercept) -0.169446   0.119837  -1.414    0.159    
## x2           0.024067   0.004347   5.537 8.28e-08 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.3946 on 233 degrees of freedom
## Multiple R-squared:  0.1163, Adjusted R-squared:  0.1125 
## F-statistic: 30.66 on 1 and 233 DF,  p-value: 8.276e-08