Shiny Application & Reproducible Pitch

This is the presentation pitch for an app created as part on the final project for the Developing Data Products course project in the Data Science specialization track. The course consists of 2 main parts:

• Shiny app with supporting documentation
• A presentation made through Rstudio Presentation to pitch the shiny app.
• The app itself produces a prediction of child's heights given the parents heights and gender based previously collected parent/child height data.
  • URL to app: https://myfapp.shinyapps.io/DEVDATAPRODUCTS/
  • URl to The source code for ui.R and server.R files: https://github.com/steveduboi/steveduboi3.github.io.git

library(UsingR)
data(GaltonFamilies)
sum1 <- summary(GaltonFamilies)
print(sum1)

     family        father         mother      midparentHeight
 185    : 15   Min.   :62.0   Min.   :58.00   Min.   :64.40  
 066    : 11   1st Qu.:68.0   1st Qu.:63.00   1st Qu.:68.14  
 120    : 11   Median :69.0   Median :64.00   Median :69.25  
 130    : 11   Mean   :69.2   Mean   :64.09   Mean   :69.21  
 166    : 11   3rd Qu.:71.0   3rd Qu.:65.88   3rd Qu.:70.14  
 097    : 10   Max.   :78.5   Max.   :70.50   Max.   :75.43  
 (Other):865                                                 
    children         childNum         gender     childHeight   
 Min.   : 1.000   Min.   : 1.000   female:453   Min.   :56.00  
 1st Qu.: 4.000   1st Qu.: 2.000   male  :481   1st Qu.:64.00  
 Median : 6.000   Median : 3.000                Median :66.50  
 Mean   : 6.171   Mean   : 3.586                Mean   :66.75  
 3rd Qu.: 8.000   3rd Qu.: 5.000                3rd Qu.:69.70  
 Max.   :15.000   Max.   :15.000                Max.   :79.00  

The app is a simple and straight forward in terms of structure and use.

You see basically 2 columns. The left column for inputing the father and mother heights, and gender, needed for the app to calculate a child heigh prediction. This done by moving 2 slider items to the desired height to explore. Do not forget to click on “Submit Button”, start app calculations.

The right colunm displays of summary the child height prediction at the top, and also a barplot the showing the the children predicted heigh vis-a-si the parents. Both items reactively change eacht time one enters input and clicks the submit button - don't forget.

As one can see from the plot (with regression line) and the summary table above, there is a positive correlation between child height and parent heights Also, predictons of child prediction is not necssarily intuitive as a tall father does not always mean a taller son.. This plot may also be an exmaple of reression to the mean.

Galton's notebook, http://www.medicine.mcgill.ca/epidemiology/hanley/galton/notebook/, transcribed by Beverley Shipley in 2001.

References

Galton, F. (1886). Regression Towards Mediocrity in Hereditary Stature Journal of the Anthropological Institute, 15, 246-263