Add alt text for exercise figures in Chapters 1 and 2

Author: DavidDiez

ch_intro_to_data/TeX/case_study_using_stents_to_prevent_strokes.tex

@@ -36,7 +36,7 @@
 \end{minipage}
 \begin{minipage}[c]{0.27\textwidth}
 \begin{center}
-\includegraphics[width = 0.75\textwidth]{ch_intro_to_data/figures/eoce/migraine_and_acupuncture_intro/earacupuncture.pdf}
+\Figures[An ear is show, with an "M" shown near the front lower lobe of the ear and an "S" shown near the middle upper portion of the ear.]{0.75}{eoce/migraine_and_acupuncture_intro}{earacupuncture}
 \end{center}
 \end{minipage}
 \begin{minipage}[c]{0.25\textwidth}

ch_intro_to_data/TeX/data_basics.tex

@@ -152,7 +152,7 @@
 \end{minipage}
 \begin{minipage}[c]{0.2\textwidth}
 \begin{center}
-\includegraphics[width = \textwidth]{ch_intro_to_data/figures/eoce/fisher_irises/irisversicolor.jpg}
+\Figures[Photo of a purple iris flower.]{}{eoce/fisher_irises}{irisversicolor}
 \end{center}
 \end{minipage}
 \begin{minipage}[c]{0.01\textwidth}
@@ -206,7 +206,7 @@
 This visualization was constructed based on a dataset where
 each observation is an airport.\footfullcite{data:usairports}
 \begin{center}
-\includegraphics[width = 0.9\textwidth]{ch_intro_to_data/figures/eoce/airports/airports.png}
+\Figures[Four copies of a map of the United States are shown in a 2-by-2 grid. For each map, the axis labels are longitude (130 degrees west to 60 degrees west) and latitude (20 degrees north to 50 degrees north). The first column of plots is labeled "private use" and the second column "public use". The first row of plots is labeled "privately owned" and the second is labeled "publicly owned". Points are shown on each of the four plots, where each point represents an airport. There appear to be many thousands of points shown in the upper-left map (private use, privately owned) and the lower-right map (public use, publicly owned), while there are relatively fewer points -- even if still numbering in the hundreds or low thousands -- in the other two plots. In all plots, there is a greater density of points shown in the Middle and Eastern portions of the United States, with a more sparse number of points over the mountain and desert areas, and then a higher concentration of points again around the states bordered by the Pacific Ocean, especially near large cities.]{0.9}{eoce/airports}{airports}
 \end{center}
 \begin{parts}
 \item
@@ -229,7 +229,7 @@
 each issue. This visualization was constructed based on a dataset where each 
 observation is a country/year pair.\footfullcite{data:unvotes}
 \begin{center}
-\includegraphics[width = 0.9\textwidth]{ch_intro_to_data/figures/eoce/unvotes/unvotes.png}
+\Figures[A grid of scatter plots with overlaid trend lines for each of three groups of points (colored green, blue, and red) per plot is shown. The grid of plots has 2 rows and 3 columns, and the plots in this description will be referenced by number, where the numbering runs from 1 to 3 in the first row and 4 to 6 in the second row. For all plots, the horizontal axis is for "year" (about 1945 to about 2018) and the vertical axis is for "percent yes" with values ranging from 0\% to 100\%. Each of the six plots summarizes voting patterns in response to a different topic at the UN General Assembly and for the countries Canada (blue), Mexico (green), and the United States (red). Each plot has points and flexible (nonlinear) trend lines fit to those points. In all cases except Plot 2 for "Colonialism", the points (data) are relatively sparse in 1940 to 1960 relative to later years. Plot 1 represents "Arms control and disarmament", which for all countries starts out low, between 0\% and 25\%, and then quickly rises by 1960 to between 25\% to 95\%, where the US remains the lowest (hovering around 25\% to 40\%), Canada a bit higher between 50\% to 70\%, and Mexico the highest and typically between 85\% to 100\%. Plot 2 is labeled "Colonialism", and the trend lines start out between 50\% to 80\%, with the US then descending close to 0\% by 1980, while Canada fluctuates between 25\% to 60\% over the duration, and Mexico rises to close to 100\% by 1980. Plot 3 represents "Economic development", where the three countries al start near 25\% to 40\%, with the US declining to about 5\% by 1990 before rising up to 20\%, Canada descending to about 25\% by 1985 before rising to 50\% by 2000 before descending again to 25\%, and Mexico rising to about 100\% by 1980 before descending to about 85\%. Plot 4 represents "Human rights", with all countries being clustered near 65\% in 1945, then the US descends to 25\% by 1975 and fluctuates between 10\% and 30\% for the rest of the time, Canada slowly descends over time to about 15\%, and Mexico rises to close to 100\% by 1985 and then descends slowly to about 80\%. Plot 5 represents "Nuclear weapons and materials", with all countries starting near 0\% in 1945, with the US then rising a bit but generally fluctuating between 15\% to 40\%, Canada rising to about 60\% by 1965 before descending to and fluctuating around 40\% to 50\%, and Mexico rising rapidly to about 90\% by 1970 then approaching 100\% over time. Plot 6 represents the "Palestinian conflict", where the countries all start between 50\% to 75\%, with the US declining steadily to about 10\% by 1985 and then approaching 5\% after that, Canada declines a bit to about 35\% in 1970 before rising to about 70\% in 2000 and then descending rapidly to close to 0\%, and Mexico gradually increases to about 95\% in 1985 and then holds roughly steady.]{0.9}{eoce/unvotes}{unvotes}
 \end{center}
 \begin{parts}
 \item List the variables used in creating this visualization.

ch_intro_to_data/TeX/review_exercises.tex

@@ -37,7 +37,7 @@
 \end{minipage}
 \begin{minipage}[c]{0.53\textwidth}
 \begin{center}
-\includegraphics[width = 0.95\textwidth]{ch_intro_to_data/figures/eoce/seattle_pet_names/seattle_pet_names.pdf}
+\Figures[A scatterplot is shown, where each point is labeled with a pet name. The horizontal axis represents "Proportion of cats" and runs from 0.002 to 0.010. The vertical axis represents "Proportion of dogs" and runs from 0.002 to 0.010. There is also a diagonal line (y = x), and only two points fall below this line: "Oliver" at about (0.0045, 0.004) and "Lily" at about (0.005, 0.004). There is a slightly positive trend in the data, the most extreme cases (highest proportions for dogs or cats) are "Lucy" at (0.006, 0.0095), "Charlie" at (0.005, 0.009), "Luna" at (0.0065, 0.007), and "Bella" at (0.005, 0.007).]{0.95}{eoce/seattle_pet_names}{seattle_pet_names}
 \end{center}
 \end{minipage}
 }{}
@@ -150,7 +150,7 @@
 \end{minipage}
 \begin{minipage}[c]{0.55\textwidth}
 \begin{center}
-\includegraphics[width = 0.78\textwidth]{ch_intro_to_data/figures/eoce/county_income_education/county_income_education_scatterplot.pdf}
+\Figures[A scatterplot is shown, with "Percent with Bachelor's Degree" on the horizontal axis (running 0\% to 80\%) and "Per Capita Income" on the vertical axis (running \$0 to \$65,000). Many thousands of points are shown. For those points with Percent with Bachelor's Degree between 0\% to 20\%, the points typically lie between the vertical ranges of \$10,000 and \$25,000. For those between 20\% to 40\% on the horizontal, the points lie mostly between \$15,000 and \$35,000 on the vertical. For those between 40\% and 60\%, the points mostly lie between \$25,000 and \$45,000. There are only about 5 points with percentages larger than 60\%, and these all lie above \$45,000 on the vertical.]{0.78}{eoce/county_income_education}{county_income_education_scatterplot}
 \end{center}
 \end{minipage}
 }{}

ch_intro_to_data/TeX/sampling_principles_and_strategies.tex

@@ -160,7 +160,8 @@
 \end{parts} \vspace{15mm}
 \end{minipage}
 \begin{minipage}[r]{0.55\textwidth}
-\hfill\includegraphics[width = 0.87\textwidth]{ch_intro_to_data/figures/eoce/internet_life_expectancy/internet_life_expectancy}
+\hfill%
+\Figures[Scatterplot with "percent of internet users" (0\% to 100\%) along the horizontal axis and "life expectancy at birth" (50 to 90) along the vertical axis. For 0\% to 15\%, about 100 points are evenly spread between 50 and 75. Then for 15\% to 90\%, the points are concentrated between about 70 and 85, and a slight upward trend is evident.]{0.87}{eoce/internet_life_expectancy}{internet_life_expectancy}
 \end{minipage}
 }{}
 

ch_summarizing_data/TeX/case_study_malaria_vaccine.tex

@@ -44,7 +44,7 @@
 \end{subparts}
 \end{minipage}
 \begin{minipage}[c]{0.5\textwidth}
-\includegraphics[width = \textwidth]{ch_summarizing_data/figures/eoce/randomization_avandia/randomization_avandia.pdf} \\
+\Figures[A histogram is shown for "Simulated rosiglitazone cardiovascular events", where values range between 2250 to 2450. The histogram, starting from the left, starts with bins that have low values until about 2280, at which point the bins rises gradually until rising steeply starting at 2320 to a peak at about 2360. The bins decline sharply at about 2380 to about half of the height of the peak, and then gradually decline out to 2460 before being zero after that point.]{}{eoce/randomization_avandia}{randomization_avandia} \\
 \end{minipage}
 }{}
 
@@ -64,8 +64,8 @@
 variable called \texttt{survived} was used to indicate whether or not the patient 
 was alive at the end of the study. \footfullcite{Turnbull+Brown+Hu:1974}
 \begin{center}
-\includegraphics[width= 0.48\textwidth]{ch_summarizing_data/figures/eoce/randomization_heart_transplants/randomization_heart_transplants_mosaic.pdf}
-\includegraphics[width= 0.48\textwidth]{ch_summarizing_data/figures/eoce/randomization_heart_transplants/randomization_heart_transplants_box.pdf}
+\Figures[A mosaic plot for variables "experiment group" (primary split) and "survived". The first tall rectangle for the "control" experiment group is about half the width of the second tall rectangle for "treatment". When looking at the secondary split for the control group, the "alive" outcome represents about 10\% of the height and "dead" represents about 90\% of the height. When looking at the secondary split for the treatment group, the "alive" outcome represents about 35\% of the height and "dead" represents about 65\% of the height.]{0.48}{eoce/randomization_heart_transplants}{randomization_heart_transplants_mosaic}
+\Figures[A side-by-side box plot is shown for the variable "Survival Time (days)" for two box plots labeled "control" and "survived". The axis for survival time spans 0 to about 1800. The box for the control group spans about 0 to 50 with the median line at about 20, and the whiskers extend down to 0 and up to about 125. There are five observations shown beyond the upper whisker at locations of about 150, 250, 300, 400, and 1400. The box for the treatment spans about 100 to 650 with the median line about 250, and the whiskers extend down to 0 and up to about 1400. There are a few points beyond the upper whiskers at about 1550, 1575, and 1800.]{0.48}{eoce/randomization_heart_transplants}{randomization_heart_transplants_box}
 \end{center}
 \begin{parts}
 \item Based on the mosaic plot, is survival independent of whether or not the 
@@ -99,6 +99,6 @@
 \end{subparts}
 \end{parts}
 \begin{center}
-\includegraphics[width= 0.6\textwidth]{ch_summarizing_data/figures/eoce/randomization_heart_transplants/randomization_heart_transplants_rando.pdf}
+\Figures[A stacked dot plot is shown for what appears to be about 100 points on the variable "Simulated Differences in Proportions", which spans values of -0.25 to 0.25. There are 11 stacks of points, which are located at the following locations and in the following approximate quantities: 2 points at -0.23, 1 point at -0.19, 8 at -0.14, 15 points at -0.10, 18 points at -0.05, 20 points at -0.01, 12 points at 0.04, 10 points at 0.08, 6 points at 0.12, 4 points at 0.17, and 3 points at 0.21.]{0.6}{eoce/randomization_heart_transplants}{randomization_heart_transplants_rando}
 \end{center}
 }{}

ch_summarizing_data/TeX/considering_categorical_data.tex

@@ -8,8 +8,8 @@
 antibiotic use in treatment of tracheitis, which is an upper respiratory 
 infection.
 \begin{center}
-\includegraphics[width = 0.45\textwidth]{ch_summarizing_data/figures/eoce/antibiotic_use_children/antibiotic_use_children_bar}
-\includegraphics[width = 0.45\textwidth]{ch_summarizing_data/figures/eoce/antibiotic_use_children/antibiotic_use_children_pie}
+\Figures[A bar plot is shown, where values on the axis range of relative frequency from 0 to just over 0.35. The values, in decreasing order and their approximate values, are Prematurity at 0.36, Cardiovascular at 0.17, Respiratory at 0.14, Trauma at 0.11, and Neuromuscular at 0.11, Genetic/metabolic at 0.07, Immunocompromised at 0.02, and Gastrointestinal at 0.02.]{0.45}{eoce/antibiotic_use_children}{antibiotic_use_children_bar}
+\Figures[A pie chart is shown of the same data from a previous chart, which was a bar chart. The Prematurity category appears to represent about a third of the pie chart (though this and other proportions are difficult to estimate accurately). The Cardiovascular group is roughly one-sixth of the total pie. About a quarter of the pie consists of an even split between Respiratory and Trauma. The remaining categories represent just under a quarter of the pie: Neuromascular (about an eighth of the pie), Genetic/metabolic (about one-fifteenth of the pie), and the remainder evenly distributed between Immunocompromised and Gastrointestinal.]{0.45}{eoce/antibiotic_use_children}{antibiotic_use_children_pie}
 \end{center}
 \begin{parts}
 \item What features are apparent in the bar plot but not in the pie chart?
@@ -65,7 +65,7 @@
 political ideology appear to be independent? Explain your reasoning.
 \footfullcite{survey:immigFL:2012}
 \begin{center}
-\includegraphics[width = 0.8\textwidth]{ch_summarizing_data/figures/eoce/dream_act_mosaic/dream_act_mosaic.pdf}
+\Figures[A mosaic plot is shown. The square (or, more accurately, a rectangle in this case), is divided into three main categories as tall rectangles: Conservative (about 40\% of the data), Moderate (about 40\% of the data), and Liberal (about 20\%). The tall rectangles are each divided into "Support", "Not Support", and "Not Sure".  The "Support" category is about 45-50\% for the Conservative and Moderate political groups and about 60-65\% for Liberal. The "Not Support" category is about 40-45\% for the Conservative and Moderate groups, while it is about 30\% for the Liberal group. In all three of the main groupings, "Not sure" is about the same, representing about 5-10\% of each political categories.]{0.8}{eoce/dream_act_mosaic}{dream_act_mosaic}
 \end{center}
 }{}
 
@@ -79,6 +79,6 @@
 political affiliation appear to be independent? Explain your reasoning.
 \footfullcite{survey:raiseTaxes:2015}
 \begin{center}
-\includegraphics[width = 0.75\textwidth]{ch_summarizing_data/figures/eoce/raise_taxes_mosaic/raise_taxes_mosaic.pdf}
+\Figures[A mosaic plot is shown for variables of political affiliation (main variable split) and opinion on whether to raise taxes on the rich, poor, or not sure. The political split, representing the main vertical splits in the mosaic plot, is roughly evenly split between Democrat, Republican, and Independent/Other, with perhaps a little more respondents in the Democrat group. The very large portion of the Democrat group -- about 85\% -- overwhelmingly supports raising taxes on the rich, with only about 5\% of this group supports raising taxes on the poor, and 5\% are unsure. About 45-50\% of the Republican and Independent/Other groups each support raising taxes on the rich, about 10\% of these groups support raising taxes on the poor, and about 40-45\% of each of these groups are not sure.]{0.75}{eoce/raise_taxes_mosaic}{raise_taxes_mosaic}
 \end{center}
 }{}