Introduction to R for Excel Users

Download the PDF

As the saying goes, when all you have is a hammer, everything looks like a nail. Excel was designed to do simple financial analyses and to craft financial statements. Though its capabilities have been expanded over the years, it was never designed to perform the sort of data analysis that industry scientists, engineers and Six Sigma belts need to perform on a daily basis.

Most data analyses performed in Excel look more like simple financial spreadsheets rather than actual data analysis, and this quality of work translates into bad—or at least sub-optimal—business decisions. There are alternatives to Excel, and the free, open-source data analysis platform R is one of them.

Unfortunately, R has a steep learning curve. I’m offering, for free, a short primer on R [PDF] where I’ve sought to make that learning curve a little less painful for engineers and scientists who normally work in Excel.


A couple of years ago, I was developing a short course to teach R to scientists and engineers in industry who normally used Excel. The goal was to help them transition to a more capable tool. My course design notes morphed into a handout, and when plans for the course fell through, that handout grew into a self-study guide, which I later adapted into this seventy-page, stand-alone introduction for Excel users.


The primer walks the reader through the basics of R, starting with a brief overview of capabilities, then diving into installation, basic operations, graphical analysis and basic statistics. I believe that a picture is worth a thousand words, so it’s light on text and heavy on examples and visuals.

The end of the book rounds out with a look at some of the most useful add-ons, the briefest of introductions to writing your own, custom functions in R, and a cross-reference of common Excel functions with their equivalents in R.

The text is broken up into chapters and fully indexed so that it can be used either as a walk-through tutorial or as a quick reference.

The qcc individuals chart as implemented using ggplot2 and grid graphics.

Update to plot.qcc using ggplot2 and grid

Two years ago, I blogged about my experience rewriting the plot.qcc() function in the qcc package to use ggplot2 and grid. My goal was to allow manipulation of qcc’s quality control plots using grid graphics, especially to combine range charts with their associated individuals or moving range charts, as these two diagnostic tools should be used together. At the time, I posted the code on my GitHub.

I recently discovered that the update to ggplot2 v2.0 broke my code, so that attempting to generate a qcc plot would throw an obscure error from someplace deep in ggplot2. The fix turned out to be pretty easy. The original code used aes_string() instead of aes() because of a barely-documented problem of calling aes() inside a function. It looks like this has been quietly corrected with ggplot2 2.0, and aes_string() is no longer needed for this.

The updated code is up on GitHub. As before, load the qcc library, then source() qcc.plot.R. For the rest of the current session, calls to qcc() will automatically use the new plot.qcc() function.

Explorable, multi-tabbed reports in R and Shiny

Matt Parker recently showed us how to create multi-tab reports with R and jQuery UI. His example was absurdly easy to reproduce; it was a great blog post.

I have been teaching myself Shiny in fits and starts, and I decided to attempt to reproduce Matt’s jQuery UI example in Shiny. You can play with the app on, and the complete project is up on Github. The rest of this post walks through how I built the Shiny app.


It’s a demo

The result demonstrates a few Shiny and ggplot2 techniques that will be useful in other projects, including:

  • Creating tabbed reports in Shiny, with different interactive controls or widgets associated with each tab;
  • Combining different ggplot2 scale changes in a single legend;
  • Sorting a data frame so that categorical labels in a legend are ordered to match the position of numerical data on a plot;
  • Borrowing from Matt’s work,
    • Summarizing and plotting data using dplyr and ggplot2;
    • Limiting display of categories in a graph legend to the top n (selectable by the user), with remaining values listed as “other;”
    • Coloring only the top n categories on a graph, and making all other categories gray;
    • Changing line weight for the top n categories on a graph, and making;

Obtaining the data

As with Matt’s original report, the data can be downloaded from the CDC WONDER database by selecting “Data Request” under “current cases.”

To get the same data that I’ve used, group results by “state” and by “year,” check “incidence rate per 100,000” and, near the bottom, “export results.” Uncheck “show totals,” then submit the request. This will download a .txt tab-delimited data file, which in this app I read in using read_tsv() from the readr package.

Looking at Matt’s example, his “top 5” states look suspiciously like the most populous states. He’s used total count of cases, which will be biased toward more populous states and doesn’t tell us anything interesting. When examining occurrences—whether disease, crime or defects—we have to look at the rates rather than total counts; we can only make meaningful comparisons and make useful decisions from an examination of the rates.


As always, we need to load our libraries into R. For this example, I use readr, dplyr, ggplot2 and RColorBrewer.

The UI

The app generates three graphs: a national total, that calculates national rates from the state values; a combined state graph that highlights the top n states, where the user chooses n; and a graph that displays individual state data, where the user can select the state to view. Each goes on its own tab.

ui.R contains the code to create a tabset panel with three tab panels.

  tabPanel("National", fluidRow(plotOutput("nationPlot"))),
  tabPanel("By State",
                      sliderInput(inputId = "nlabels",
                                  label = "Top n States:",
                                  min = 1,
                                  max = 10,
                                  value = 6,
                                  step = 1)
  tabPanel("State Lookup",

Each panel contains a fluidRow element to ensure consistent alignment of graphs across tabs, and on tabs where I want both a graph and controls, fluidRow() is used to add the controls below the graph. The controls are placed inside a wellPanel() so that they are visually distinct from the graph.

Because I wanted to populate a selection menu (selectInput()) from the data frame, I created the selection menu in server.R and then displayed it in the third tab panel set using the htmlOutput() function.

The graphs

The first two graphs are very similar to Matt’s example. For the national rates, the only change is the use of rates rather than counts.

df_tb <- read_tsv("../data/OTIS 2013 TB Data.txt", n_max = 1069, col_types = "-ciiii?di")

df_tb %>%
  group_by(Year) %>%
  summarise(n_cases = sum(Count), pop = sum(Population), us_rate = (n_cases / pop * 100000)) %>%
  ggplot(aes(x = Year, y = us_rate)) +
  geom_line() +
  labs(x = "Year Reported",
       y = "TB Cases per 100,000 residents",
       title = "Reported Active Tuberculosis Cases in the U.S.") +

The main trick, here, is the use of dplyr to summarize the data across states. Since we can’t just sum or average rates to get the combined rate, we have to sum all of the state counts and populations for each year, and add another column for the calculated national rate.

To create a graph that highlights the top n states, we generate a data frame with one variable, State, that contains the top n states. This is, again, almost a direct copy of Matt’s code with changes to make the graph interactive within Shiny. This code goes inside of the shinyServer() block so that it will update when the user selects a different value for n. Instead of hard-coding n, there’s a Shiny input slider named nlabels. With a list of the top n states ordered by rate of TB cases, df_tb is updated with a new field containing the top n state names and “Other” for all other states.

top_states <- df_tb %>%
      filter(Year == 2013) %>%
      arrange(desc(Rate)) %>%
      slice(1:input$nlabels) %>%

df_tb$top_state <- factor(df_tb$State, levels = c(top_states$State, "Other"))
df_tb$top_state[$top_state)] <- "Other"

The plot is generated from the newly-organized data frame. Where Matt’s example has separate legends for line weight (size) and color, I’ve had ggplot2 combine these into a single legend by passing the same value to the “guide =” argument in the scale_XXX_manual() calls. The colors and line sizes also have to be updated dynamically for the selected n.

    df_tb %>%
      ggplot() +
      labs(x = "Year reported",
           y = "TB Cases per 100,000 residents",
           title = "Reported Active Tuberculosis Cases in the U.S.") +
      theme_minimal() +
      geom_line(aes(x = Year, y = Rate, group = State, colour = top_state, size = top_state)) +
      scale_colour_manual(values = c(brewer.pal(n = input$nlabels, "Paired"), "grey"), guide = guide_legend(title = "State")) +
      scale_size_manual(values = c(rep(1,input$nlabels), 0.5), guide = guide_legend(title = "State"))


The last graph is nearly a copy of the national totals graph, except that it is filtered for the state selected in the drop-down menu control. The menu is as selectInput() control.

    selectInput(inputId = "state", label = "Which state?", choices = unique(df_tb$State), selected = "Alabama", multiple = FALSE)

With a state selected, the data is filtered by the selected state and TB rates are plotted.

df_tb %>%
  filter(State == input$state) %>%
  ggplot() +
  labs(x = "Year reported",
       y = "TB Cases per 100,000 residents",
       title = "Reported Active Tuberculosis Cases in the U.S.") +
  theme_minimal() +
  geom_line(aes(x = Year, y = Rate))

Wrap up

I want to thank Matt Parker for his original example. It was well-written, clear and easy to reproduce.

Sample Size Matters: Design and Cost


  • R Core Team (2014). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL
  • H. Wickham. ggplot2: elegant graphics for data analysis. Springer New York, 2009.

Sample Size Matters: Design and Experiments

Previously, I introduced the idea that samples do not look exactly like the populations that they are drawn from, and had a closer look at what impact sample size has on our ability to estimate population statistics like mean, proportion or Cpk from samples. Here, I will have a closer look at how this uncertainty impacts our engineering process. In the next post, I will tie in the engineering impacts and decisions to the business value and costs.

Difference to detect

When we are testing, we’re either testing to determine that the new product or process performs better than the old or, for cost reduction projects, that the cheaper product or process is at least as good as the existing one.

This means that we need to detect a difference between old and new values, such as a difference in the mean weight between the new and old parts. The larger the sample size, n the smaller the difference, \Delta that we can detect. The error, \epsilon, in our estimate of the differences gets smaller as sample size increases:

Given the uncertainties in our estimate of \mu and \sigma, illustrated above, it should be clear now that with small sample sizes we can only detect large differences of many multiples of the sample standard deviation, S.


When trying to determine if a new product or process is better than an old one, we are usually interested in shifting the mean. We want a product to be lighter, provide more power, or a process to work faster. In such cases, we need to estimate the difference of the means, \Delta = \mu_2 - \mu_1 and ensure that it is different than 0 (or some other pre-determined value). The minimum difference that we can reliable detect is plotted below for different sample sizes.

Minimum difference to detect in means by sample size

Standard deviation

In many Six Sigma projects, and any time we want to shift the mean closer to a specification limit, we need to compare the new population standard deviation with the old. The simplest way of making this comparison is by taking the ratio F = \sigma_{2}^{2} / \sigma_{1}^{2}, where \sigma_{2}^{2} is the larger of the two variances. The dependence on sample size is illustrated below.

Minimum difference detectable in standard deviations by sample size

You can see from the inset plot, which includes sample sizes of 2 and 3, that small sample sizes really hurt comparisons of variance, and that interesting differences in variance can’t be detected until we have more than 10 samples.


Proportions, such as fraction of defective parts between a new and old design, can be compared by looking at the difference between the two proportions, \Delta = \left| p_1 - p_0 \right|.

Minimum difference detectable in proportions by sample size

You can see from this that proportions data provides much less information than variable data; we need much larger sample sizes to achieve usefully small \Delta.

Summary and look forward

When designing experiments, the goal is to detect some difference between two populations. The uncertainty in our measurements and the variation in the parts has a big impact on how many parts we need to test, or greatly limits what we can learn from an experiment.

Next time, I’ll show how these calculations of sample size and uncertainty impact the busines.

Sample Size Matters: Uncertainty in Measurement

In my previous post, I gave a brief introduction to populations and samples, and stated that sample size impacts our ability to know what a population really looks like. In this post, I want to show this relationship in more detail. In future posts, I will look at how sample size considerations impact our engineering process and what impacts this has on the business.

Mean and sample size

The error in our estimate of the mean, E, is proportional to the standard deviation of the sample, S, and the sample size, n.

We can visualize this easily enough by plotting the 95% confidence interval. When we sample and calculate the sample mean (\overline{X}), the true population mean, \mu, (what we really want to know) is likely to be anywhere in the shaded region of the graph below.

Uncertainty in mean

This graph shows the 95% confidence region for the true population mean, \mu; there’s a 95% chance that the true population mean is within this band. The “0” line on the y axis is our estimate of the mean, \overline{X}. We can’t know what the true population mean is, but it’s clear that if we use more samples, we can be sure that our estimate is closer to the true mean.

Standard deviation and sample size

Likewise, when we calculate the sample standard deviation, S, the true standard deviation, \sigma has a 95% chance of being within the confidence band below. For small sample sizes (roughly less than 10), the measured standard deviation can be off from the true standard deviation by several times. Even for ten samples, the potential error is nearly \pm 1 standard deviation.

Uncertainty in standard deviation

Proportion and sample size

For proportions, the situation is similar: there is a 95% chance that the true sample proportion, p, is within the shaded band based on the measured sample proportion \hat{p}. Since this confidence interval depends on \hat{p} and cannot be standardized the way \mu and \sigma can be, confidence intervals for two different proportions are plotted.

Uncertainty in proportion

For small n, proportions data tells us very little.

Process capability and production costs

The cost of poor quality in product or process design can be characterized by the Cpk:

Where USL is the upper specification limit (also called the upper tolerance) and LSL is the lower specification limit (or lower tolerance).

We can estimate the defect rate (defects per opportunity, or DPO) from the Cpk:

DPO = 1 - \Pr\left(X < 3 \times Cpk - 1.5\right)

That probability function is calculated in R with pnorm(3 * Cpk - 1.5) and in Excel with NORMSDIST(3 * Cpk - 1.5). The 1.5 is a typical value used to account for uncorrected or undetected process drift.

Since we don’t know \mu and \sigma, we have to substitute \overline{X} and S. The uncertainty in these estimates of the population \mu and \sigma mean that we have uncertainty in what the true process Cpk (or defect rates) will be once we’re in production. When our sample testing tells us that the Cpk should be 1.67 (the blue line), the true process Cpk will actually turn out to be somewhere in the shaded band:

Uncertainty in Cpk

Below the blue line, our product or process is failing to meet customer expectations, and will result in lost customers or higher warranty costs. Above the blue line, we’ve added more cost to the production of the product than we need to, reducing our gross profit margin. Since that gray band doesn’t completely disappear, even at 100 samples, we can never eliminate these risks; we have to find a way to manage them effectively.

The impact of this may be more evident when we convert from Cpk to defect rates (ppm):

Uncertainty in ppm

Summary and a look forward

With a fair sampling process, samples will look similar to—and statistically indistinguishable from—the population that they were drawn from. How much they look like the population depends critically on how many samples are tested. The uncertainties, or errors in our estimates, resulting from sample size decisions have impacts all through our design analysis and production planning.

In the next post, I will explore in more detail how these uncertainties impact our experiment designs.

Sample Size Matters

I find that Six Sigma and Design for Six Sigma courses are often eye-opening experiences for participants. There is an experience of discovering that there are tools available to answer problems that have vexed them, and learning that good engineering and science decisions can lead directly to good business outcomes through logical steps.

One of the most remarkable such moments is when students realize the importance of sample size. In the best cases, there is a forehead-slapping moment where the student realizes that much of the testing they’ve done in the past has probably been a complete waste of time; that while they thought they were seeing interesting differences and making good decisions, they were in fact only fooling themselves by comparing too-small data sets.

I want to show in the next few blog posts why sample size matters, both from a technical perspective and from a business perspective.

Design example

Throughout the next few posts, I’ll use the example of a manufactured product which the customer requires weigh at least 100 kg, sells for about $140 and that costs $120 to manufacture and convert to a sale (the cost of goods sold, or COGS, is $120).

Sales 140
COGS 120
Material 60
Labor and Overhead 60
Gross Profit 20

We want to develop a new version of the product, using a modified design and a new process that, by design, will reduce the cost of material by 10%. The old cost of material was 50% of COGS, or $60. To achieve the material cost reduction of 10%, we have to remove $6 in material costs, improving gross profit to $26.

We believe that the current design masses 120 kg, so we estimate that our new part mass should be 120 - 0.1 \times 120 = 108 kg.

Current Design New Design Target
Part Weight 120 108

Seems like we might be done at this point, and I’ve seen plenty of engineering projects that stop here. Unfortunately, this isn’t the whole story. Manufacturing will be unable to produce parts of exactly 108 kg, so they’ll need a tolerance range to check parts against. We have that customer requirement for at least 100 kg, so any variation has to stay above that. We also want to save money relative to the current design, so we don’t want many parts to weigh much more than this, especially since the customer isn’t really willing to pay us for the “extra” material beyond 100 kg.

Population versus sample statistics

Most of process or product improvement is concerned with reducing the standard deviation, \sigma, shifting the mean (a.k.a. average), \mu, or reducing a proportion, p, of a process or product characteristic. These summary statistics refer to the population characteristics—the mean, standard deviation or proportion of all parts of a certain design that will ever be produced, or all times that a production step will ever be completed in the intended manner.

Since we can’t measure the whole population up front—we will be producing parts for a long time—we have to draw a sample from the population, and use the statistics of that sample to gain insight into the total population. We can visualize this, somewhat crudely, with the following:

Sampling from known population

We can imagine that the blue circles are conforming parts, and the orange octagons are non-conforming parts. If the sampling process is fair, then the sample proportion \hat{p} will be close to—and statistically indistinguishable from—the true population proportion p. In the population we have 44 parts total, 8 defective parts and 36 conforming parts. In the sample that we drew, we have 10 parts total, 9 conforming and 1 defective. While (p = 8/36 = 1/4 \ne \hat{p} = 1/9, statistically we have

matrix(c(1, 8, 10-1, 44-8), ncol=2) %&gt;% 
  chisq.test(simulate.p.value = TRUE)
##  Pearson's Chi-squared test with simulated p-value (based on 2000
##  replicates)
## data:  matrix(c(1, 8, 10 - 1, 44 - 8), ncol = 2)
## X-squared = 0.3927, df = NA, p-value = 0.6692

With such a high p-value (0.67), we fail to reject the null hypothesis that \hat{p} = p; in more colloquial terms, we conclude that the apparent difference between 8/36 and 1/9 is only due to random errors in sampling. (For larger counts of successes and failures, prop.test() would also work and would be more informative.)

From our perspective, of course, we don’t know what the population looks like. We don’t have any way of knowing with certainty—or accessing data about—future performance, so there is no way for us to know what the total population looks like. In lieu of population data, we develop a sampling process that allows us to fairly draw a sample from that population.

Sampling from unknown population

While we want to know the true population mean, \mu, the true population standard deviation, \sigma, or the true population proportion p, we can only calculate the sample mean, \overline{X}, the sample standard deviation, S, or the sample proportion \hat{p}.

From the known sample, we then reason backward to what the true population looks like. This is where statistics comes into play; statistics allows us to place rigorous boundaries on what the population may look like, without fooling ourselves. Sample size is critical to controlling the uncertainty in these boundaries.

Summary and a look forward

Testing in product development—and usually in production—involves sampling a product or process. Samples never look exactly like the population that we are concerned about, but if the sampling process is fair then the samples will be statistically indistinguishable from the population. With due awareness of the statistical uncertainties, we can use samples to make decisions about the population.

In the next post, I will look at how sample size impacts the uncertainty in our estimation of population statistics like the mean and standard deviation. In a later post, I will look at how this uncertainty impacts the business.

A short aside on statistical tests for proportions

The usual way to compare two proportions would be a proportions test (prop.test() in R), but because we have so few samples to compare, the results may be unreliable and prop.test() generates an appropriate warning. fisher.test() provides an exact estimate of the p-value, but the assumptions are violated with data like this, where we are sampling a fixed number of parts (i.e. row sums are fixed, but column sums are not controlled). This leaves us with using a chi-squared test (chisq.test() in R) which is less informative but does the job. Either the Barnard test or Bayesian estimation based on Monte Carlo simulation would be more informative and possibly more robust.