In popiela/homework: Homework assignment

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(knitr)
opts_chunk$set(
    comment = "",
    #fig.width = 12, 
    message = FALSE,
    warning = FALSE,
    tidy.opts = list(
        keep.blank.line = TRUE,
        width.cutoff = 150
        ),
    options(width = 150),
    eval = TRUE,
    echo = TRUE,
    fig.height =8,
    fig.width = 6,
    fig.align = "left"
)
# print head
# don't print long-lines

library(homework)

Ordinal logistic regression coefficients

knitr::kable(
 homework::model_ordinal(experiment_data) 
)

Ordinal logistic regression was used in order to find out how each attribute influences the overall likeliness to download

Below is the interpretation for coefficients that were statistically significant.

1. For a price of $40/month, the odds of being very likely to download (vs. Somewhat likely or Somewhat unlikely or Very unlikely) the mobile app are 39% lower than for a price of $20/ month, holding constant all other variables.

2. For a price of $30/month, the odds of being very likely to download (vs. Somewhat likely or Somewhat unlikely or Very unlikely) the mobile app are 30% lower than for a price of $20/ month, holding constant all other variables.

3. For an outcome "empowering you to take back your sleep habits", the odds of being very likely to download (vs. Somewhat likely or Somewhat unlikely or Very unlikely) the mobile app are 11% lower than for an outcome "breaking bad habits and creating new routines", holding constant all other variables.

4. For a rtb "daily text messages from a coach", the odds of being very likely to download (vs. Somewhat likely or Somewhat unlikely or Very unlikely) the mobile app are 15% lower than for an outcome "a program created just for you", holding constant all other variables.

5. For social proof "scientific evidence", the odds of being very likely to download (vs. Somewhat likely or Somewhat unlikely or Very unlikely) the mobile app are 11% higher than for social proof "a method that has helped thousands", holding constant all other variables.

Counts of demographic data - interactive plot

homework::demo_plot(survey_data)

Counts of behavioral questions data - interactive plot

homework::behavioral_plot(survey_data)

K - means clustering of philosophy data

Profiling the clusters based on respondents age groups :

knitr::kable(
 homework::kmeans_cluster(survey_data) 
)

Ward's method clustering of philosophy data

knitr::kable(
 homework::ward_cluster(survey_data) 
)

Profiling the clusters based on respondents age groups and gender :

Cluster 1 - General group

Cluster 2 - Female group (less popular)

Cluster 3 - Female 46-64 age group

Cluster 4 - 31-45 age group

Cluster 5 - Female group (more popular)

Cluster 6 - 18-30 age group

Cluster 7 - 46-64 age group

Cluster 8 - Mixed group

popiela/homework documentation built on Jan. 17, 2021, 1:22 a.m.