- Information on Midterm III posted

- More practice problems for Midterm III
- t-test example:

```
library(tidyverse)
# 1. Are these 10 values significantly different from 2?
values <- data_frame(
x = c(3.192, 3.842, 5.423, 3.181, 4.317, 1.845, 3.079, 3.426, 2.561, 2.221)
)
# 2. Uhhh. Maybe? Let's visualize:
ggplot(values, aes(x = x)) +
geom_histogram(bins = 4) +
geom_vline(xintercept = 2, col="red")
```

mean_x |
---|

3.3087 |

```
# 4. Let's do this formally using the t-test!
t.test(x = values$x, alternative = "two.sided", mu = 2)
```

```
##
## One Sample t-test
##
## data: values$x
## t = 3.9756, df = 9, p-value = 0.003227
## alternative hypothesis: true mean is not equal to 2
## 95 percent confidence interval:
## 2.564043 4.053357
## sample estimates:
## mean of x
## 3.3087
```

```
# 5. Let's plot the null distribution of the t-statistic!
ggplot(data.frame(x = c(-4, 4)), aes(x = x)) +
stat_function(fun = dt, args = list(df = 9)) +
geom_vline(xintercept = 3.9756, col = "red")
```

- Exit survey posted. 2.5% of your final project grade is based on merely completing this exit survey.
- In particular questions relating to use of Slack, as at the Symposium on Data Science and Statistics (SDSS) I’ll be presenting “Using Slack for Communication and Collaboration in the Classroom”

- FYI other SDS faculty will be presenting at SDSS as well:
- Prof. Katherine Kinnaird: “Data Physicalizations: Where Art, Data, and Domain Applications Combine”
- Prof. Miles Ott: “Ethics in an Advanced Undergraduate Seminar: Statistical Analysis of Social Network Data”

- Midterm III review
- Practicing statistical inference (confidence tests & hypothesis tests) via mathematical formulas using pen & paper instead of simulations using a computer.

- Conditions for inference for regression: last element needed for term project.
- Simple regression example:

moderndive readings in schedule above.

Inference for regression. In other words sampling scenarios 5 & 6 from **moderndive Table 8.6**:

Scenario | Population parameter | Notation | Point estimate | Notation. |
---|---|---|---|---|

1 | Population proportion | \(p\) | Sample proportion | \(\widehat{p}\) |

2 | Population mean | \(\mu\) | Sample mean | \(\widehat{\mu}\) or \(\overline{x}\) |

3 | Difference in population proportions | \(p_1 - p_2\) | Difference in sample proportions | \(\widehat{p}_1 - \widehat{p}_2\) |

4 | Difference in population means | \(\mu_1 - \mu_2\) | Difference in sample means | \(\overline{x}_1 - \overline{x}_2\) |

5 | Population regression slope | \(\beta_1\) | Fitted regression slope | \(\widehat{\beta}_1\) or \(b_1\) |

6 | Population regression intercept | \(\beta_0\) | Fitted regression intercept | \(\widehat{\beta}_0\) or \(b_0\) |

Recall from Chapter 6 our study of relationship between the following two variables for instructors of \(n\) = 463 courses at the UT Austin:

- \(y\): instructor teaching score as given by students
- \(x\): instructor “beauty” score as “rated” by a panel of 6 students

Recall our exploratory data visualization of the relationship in **moderndive Figure 6.4**:

and the corresponding regression table in **moderndive Table 6.2**:

term | estimate | std_error | statistic | p_value | lower_ci | upper_ci |
---|---|---|---|---|---|---|

intercept | 3.880 | 0.076 | 50.961 | 0 | 3.731 | 4.030 |

bty_avg | 0.067 | 0.016 | 4.090 | 0 | 0.035 | 0.099 |

- I will be absent on Friday. Guest lecturer: Prof. Ben Capistrant from the Smith School of Social Work and current SDS/MTH 291 Multiple Regression instructor.

- Went over midterm II solutions.

- moderndive readings in schedule above.

- No office hours on Wednesday 4/24
- Term project resubmission instructions posted under Term Project.

None

moderndive readings in schedule above.

Project:

- All feedback sessions have held.
- Final instructions/template will be posted on Monday.
- During final lab on Tue 4/30 you’ll be working on project.
- Note for those of you who did a log10-transformation of your outcome variable.

- The 😕🤕😵🤯😱 statistical definitions, terminology, and notation for hypothesis testing.
- Remember that at the root of all hypothesis testing, just like with confidence intervals, is
**sampling**!

- The question is: in real-life where we take only one sample, how can we study the effects of sampling variation? Using resampling!
- Confidence intervals: bootstrap resampling
**with**replacement - Two-group hypothesis testing: permutation resampling
**without**replacement. i.e. shuffle it!

- Confidence intervals: bootstrap resampling

moderndive readings in schedule above.

For those of you who are Game of Thrones fans and those of you who are potentially interested in Machine Learning, check out this tweet. Season 8, Episode 1 spoiler alert!

- Discussion on census bureau checkboxes for “race” and the politics behind their determination.
- Discussion on tweet of the day.

- The intuition behind hypothesis testing.
~~(Time permitting) The statistical definitions/terminlogy behind hypothesis testing; in particular how hypothesis testing relates to sampling.~~

If you care to share feedback relating to the use of this example, please fill out this Google Form. You have the option to remain anonymous and all responses will remain confidential.

Thanks to @AmstatNews for recognizing the importance of the gender spectrum, especially wrt the statistics classroom and study design. I'm really proud of this article with Jack Miller. Also thanks to @Miles_Ott !!

— Jo Hardin (@jo_hardin47) March 1, 2019

https://t.co/jlxMoVqWir

- Lab tomorrow (Tue 4/16) is optional office hours; Jenny will be in Sabin-Reed 301.
- No problem set this week!
- Don’t forget your project feedback appointments this week.
- Talk on Thursday April 18th 6pm in Seelye 106

**RStudio Desktop users only**: reinstall/update the`moderndive`

package- Form groups of two students
- Watch slideshow
- Imagine… a hypothetical world with no gender discrimination in hiring.
- In this hypothetical world, we can switch i.e.
**shuffle**i.e.**permute**the (binary)`gender`

variable in . Do this using this code to create a new variable`hypothetical`

```
library(moderndive)
library(tidyverse)
promotions <- promotions %>%
mutate(hypothetical = sample(gender, replace = FALSE))
View(promotions)
```

- Fill in worksheet
**Results**:

- Go over practice midterm II.

- No lecture on Wednesday, extra office hours on Friday 1-3pm.
- Midterm II review. In particular practice midterm posted on Midterms page.
- Project feedback: Sign up for a feedback session next week where all group members and myself will record a screencast in my office.
**Please be mindful of how much coordination it takes for me to schedule 17 feedback sessions and read all these instructions carefully first**:- Jump to the week of April 14-20 (next week) on my Google Appointments Calendar and identify which “220 Project Feedback Only” 15 minute time slots work for you.
- Then as a group, coordinate on a “220 Project Feedback Only” 15 minute time slot
**when all group members can attend.** - Then only the group leader will book the 15 minute appointment. Under “Description” include both 1) your group name and 2) the names of all group members.
- If none of the listed times work for you all, contact me in the Slack Direct Message that includes me and Jenny and all group members.

- Project submission due at 5pm today!
- Midterm II next week. Review on Monday

- Recap of Lec 27: The “frequentist” intepretation of confidence intervals
- Two ways to compute SE.

- Note: As seen in moderndive 9.5, I still need to add
`tactile_shovel_1`

data frame to`moderndive`

package. - moderndive readings in above schedule.

- the
`infer`

bootstrapping framework (see handout below). Both the left-hand side which uses`dplyr`

verbs and the right hand side that uses`infer`

verbs do the same thing. However we’ll see that the`infer`

code on the right can be used in more situations.

- Intepreting confidence intervals
- What determines the width of net?
- The confidence level: A 95% CI will be wider than an 80% CI
- The original sample size n. As n goes up, the CI gets narrow i.e. you have more precise results.

Read the following in moderndive:

- Re-read moderndive 9.4 in light of today’s discussion
- moderndive 9.5-9.6

- No lecture today

- Recap of Lec24
- Normal distribution/bell curve theory
- moderndive 9.3: What is a confidence interval?
- moderndive 9.4: The resampling framework

```
library(tidyverse)
bootstrap_distn <- read_csv("https://rudeboybert.github.io/SDS220/static/resample_means.csv")
# New: Compute 2.5th and 97.5th percentiles
conf_int <- quantile(bootstrap_distn$sample_mean, probs = c(0.025, 0.975), na.rm = TRUE)
conf_int
```

```
## 2.5% 97.5%
## 1990.956 1998.699
```

Read the following in moderndive:

- Appendix A.2 on the Normal distribution
- moderndive 9.3-9.4

- Please open Slack.

- Recap of Lec23:
- Recall your “tactile” resampling results in this Google Sheet
- Let’s load each of your resampled sample means and plot a histogram.

- Today:
**New inference scenario**Number 2 in moderndive Table 8.6. Unknown value is no longer population proportion \(p\) but population mean \(\mu\)- Just like in Ch8 with sampling, we are going from “tactile resampling” by hand to “virtual resampling” using a computer!

Read the following in moderndive:

- Intro to Ch9
- 9.1
- Read 9.1.1
- Skip 9.1.2 and 9.1.3. These sections will be re-written before book launch
- Read 9.1.4

- 9.2

- Tomorrow 12:15-1:05pm in Ford Hall Atrium: Presentation of SDS major

- What is the
`tidyverse`

package?

- Recap of Chapter 8: Sampling. In particular two goals:
- Study the effect of sampling variation on our estimates.
- Study the effect of sample size on sampling variation.

- In real-life, when we have a single sample of size \(n\) (and not 1000 like in our simulations), what do we do? Bootstrap
**re**-sampling from the original sample!

Resampling tactile exercise. Open this Google Sheet. Resulting 35 sample means based on a resample of size 50 are here:

- Discuss next phase of project: “Project (initial) submission” due Fri 4/5 5pm.
- Talk today from 12:15-1:00pm in McConnell B15: Gina DelCorazon ’04, Director of Data & Analytics at The National Math and Science Initiative

- Discussion on sampling distribution handout.

- moderndive readings in above schedule.

- No office hours today. I do however have appointments to book on Friday (see syllabus for link).
- Project proposal feedback given today. On Friday I will post information about the next phase “Project (initial) submission” due Fri 4/5 5pm.

- Overall comments about project proposal.
- Recap of Lec20: Simulation. Goal is to study the effect of sampling variation.
- Sampling terminology, notation, and statistical definitions. Mastering these will take practice, practice, practice.

- moderndive readings in above schedule.

Relating to the formating of your reports and in particular it’s length. Do not include “superfluous” output as it only increases the “ink to information ratio.”

And be mindful of the work you leave for others

— Judge John Hodgman is supported by #maxfundrive (@hodgman) July 29, 2016

- Over spring break moderndive Chapters 6 & 7 on basic and multiple regression went through a thorough renovation, so the presentation might be a little different. Please let me know if you have comments, questions, or feedback.
- Two events coming up:
- Fri 3/22 12:15-1:00pm in McConnell B15: Gina DelCorazon ’04, Director of Data & Analytics at The National Math and Science Initiative
- Tue 3/26 12:15-1:05pm in Ford Hall Atrium: Presentation of SDS major

**Sampling exercise**:

- Ask yourself “What proportion of this bowl’s balls are red?”
- Come up to the front of the class and take a photo of the sample.
**Do not delete this photo as you’ll be submitting it later** - Compute the proportion of the 50 balls that are red.
- Post a post-it on the histogram on the blackboard where the bins are
*left-inclusive*. In other words, if you obtain a proportion of 0.2, put a post-it in the 0.2-0.25 bin.

**Why are we doing this?**

- To study the effects of
*sampling variation* - Also, to update the contents of moderndive section 8.1.

- moderndive readings in above schedule.

- Went over Midterm I

- Slack announcement + poll
- Guest lecturer today at 11:45am: Prof Randi Garcia.
- Spinelli Center tutoring hours change tomorrow (Thursday 3/7) only: 4:30-6pm and not
~~7-9pm~~. - Tweet of the day
- From New York Times article “For Big-Data Scientists, ‘Janitor Work’ Is Key Hurdle to Insights”:

Yet far too much handcrafted work, what data scientists call “data wrangling,” “data munging” and “data janitor work”, is still required. Data scientists, according to interviews and expert estimates,

spend from 50 percent to 80 percent of their timemired in this more mundane labor of collecting and preparing unruly digital data, before it can be explored for useful nuggets.

- Recap of Lec17
- More on designed experiments from OpenIntro Section 1.5 page 17 on Experiments; to access PDF of OpenIntro, click on “free online” link here.
**Blocking**in a randomized experiment:

Help Chester and me make this book as good as it can be before it goes to press! If something you read doesn’t make sense, let me know!

Good news everyone! We’ve signed with @CRCPress for a print edition with our new title: “Statistical Inference via Data Science: A moderndive into R & the tidyverse” by @old_man_chester & @rudeboybert! Coming Fall 2019! Until then check out a preview at https://t.co/X5zI0LvHGb pic.twitter.com/Wa049JBeaJ

— ModernDive (@ModernDive) March 4, 2019

- Guest lecturer on Wed 3/6 at 11:45am: Prof Randi Garcia.
- Reminder tomorrow is:

- Recap of Lec11: model selection
- Random assignment, causal inference, observational studies vs experiments.
- Example:
- Discussion questions:
- Why did I ask the question “Have you been to Africa before?”
- Why did I have the people with even-numbered birthdays take the “Africa Quiz” and those with odd-numbered birthdays take “Africa Experiment”?
- Comment on what you think the
**difference**between the two histogram for heights will be for those who took the “Africa Quiz” vs “Africa Experiment”.

- Extra office hours on Friday 2:45-4pm.

- Question: How do we choose between interaction and parallel slopes models?
- Answer: In
`multiple_regression.html`

report. You can download the`multiple_regression.Rmd`

R Markdown file as well if you are curious.

- moderndive readings in above schedule.

- Clarifications of upcoming deadlines.
- Project proposal phase posted on Term Projects page.
- Extra office hours on Friday 2:45-4pm.

- Midterm I details posted.
- Term project: I will
- Give feedback on your “data proposals” on Slack by later today.
- Post instructions for the next “Project proposal” phase by Wednesday; this phase is due Fri 3/8 5pm and involves data wrangling and exploratory data analysis.

- Multiple regression

- moderndive readings in above schedule.

**If you are working with RStudio Desktop, please follow these steps before tomorrow’s lab. If you get stuck, please ask Jenny for help.****If you are working on RStudio Server, you can ignore these steps.**

We’re going to install the development AKA beta-version of the `moderndive`

package, which includes a new function `gg_parallel_slopes()`

allowing you to create a ggplot of the **parallel slopes model**.

- Install the
`devtools`

package as you normally would install a package. Say yes to any prompts. - Run the following line in your console to install the development version of the
`moderndive`

package off of GitHub.com:

`devtools::install_github("moderndive/moderndive", ref = "geom_parallel_slopes")`

- Run
`library(moderndive)`

to reload the package. - Run
`?gg_parallel_slopes`

and see if the help file pops up. If it does, your installation worked! - Run example code at the bottom of the help file to see it in action! You should get the following plot:

- Seelye self-scheduled Midterm I next Fri 3/1 thru Sun 3/3; Midterm I review on Monday.

- Boxplots for EDA when explanatory variable \(x\) is categorical.
- Indicator function
- What are fitted values and residuals when \(x\) is categorical?

- moderndive readings in above schedule.

Dr. Benn is excited for her talk at Smith College SDS! **Are you?**

Yessss!!! Definitely looking forward to speaking with and learning from the amazing @SmithCollegeSDS faculty and students! 🤗💃🏾🤗💃🏾 https://t.co/te01giEvXv

— Emma Benn (@EKTBenn) February 21, 2019

- None

- Recap of Lec11: What do we mean by “best” fitting line? Note in the plot below there are 3 points marked with black dots along with:
- The “best” fitting regression line in blue
- An arbitrarily chosen line in dashed red
- Another arbitrarily chosen line in dashed green

- Regression using a categorical explanatory variable

```
library(dplyr)
library(moderndive)
# Example data for chalk talk
example <- tibble(
name = c("Bert", "Bert", "Bert", "Jenny", "Jenny", "Jenny", "Miles", "Miles", "Miles"),
value = c(9, 10, 11, 11, 12, 13, 8, 9, 10)
)
# Get regression table
model_example <- lm(value ~ name, data = example)
get_regression_table(model_example)
```

term | estimate | std_error | statistic | p_value | lower_ci | upper_ci |
---|---|---|---|---|---|---|

intercept | 10 | 0.577 | 17.321 | 0.000 | 8.587 | 11.413 |

nameJenny | 2 | 0.816 | 2.449 | 0.050 | 0.002 | 3.998 |

nameMiles | -1 | 0.816 | -1.225 | 0.267 | -2.998 | 0.998 |

- Work on projects.
- moderndive readings in above schedule.

- Part of next lecture (Lec12 on Wed 2/20) will be devoted to work on project.
- Note in the above schedule that the topics for Lab06 and Lab07 have switched places, as originally it erroneously had you working on your project proposals
**after**it was due.

- Recap of Lec10
- What is a confounding variable?
- Fitted values & residuals via
`get_regression_points()`

- What do we mean by “best” when we say that the regression line is the “best fitting” line?

- moderndive readings in above schedule.

- Project data is due in a week!
- We’ll devote part of Lec12 on Wed 2/20 to work on project data phase.

- Recap of Lec09
- Regression table via
`get_regression_table()`

and interpreting the regression line

- moderndive readings in above schedule.

- Winner for best group name.

- Recap of Lec08
- Correlation coefficient

- moderndive readings in above schedule.

Why are Jenny and Albert always on your cases about running `glimpse()`

and `View()`

on your data frames? **Looking at your data** is so deceptively simple that many people forget or ignore this step, even analysts/engineers with PhD’s at Google! Before performing any kind of analysis, you must getting a sense of:

**What types of variables you have in your columns?**Numerical, categorical, text, dates?**What values you have in your cells?**Units of any measurements?**What is the quality of your data?**Do you have missing data? Are there crazy outliers?

These are the most fundamental steps to take before any data analysis! That’s why moderndive starts in Chapter 2 with “Data exploration” with `glimpse()`

and `View()`

.

Shout-out to people like me who like taking a look at even huge tables of raw data before summarizing or transforming, to get a sense of what's in there and spot potential data quality issues. Step 1 in data exploration! pic.twitter.com/wVLHWcVNWA

— Data Science Renee (@BecomingDataSci) February 12, 2019

- Everybody join the
`term_project`

channel in Slack. - Discuss Data due phase of term-project.

- Recap of Lec07:
- What does
`group_by()`

by itself do? - Difference between
`filter()`

and`group_by() %>% summarize()`

- What does
- Last three verbs:
`mutate()`

existing variables to create new ones`arrange()`

rows in ascending or`desc()`

ending alphanumeric order of another variable`select()`

or drop variables

moderndive readings in above schedule.

- Term project groups are due today at 5pm. Make sure your group leader has completed all three steps, in particular the Google Form.
- I will introduce next phase of term project on Monday: Data due: Fri 2/22 5pm.

- Recap of Lec06
- Computing summary statistics using
`summarize()`

- Adding
`Groups`

*meta-data*using`group_by()`

. See example code below. - Computing summary statistics split by group using
`group_by() %>% summarize()`

moderndive readings in above schedule.

- Reminder that “Multiple Imputation Methods in Cluster Randomzied Trials” by Prof. Brittney Bailey from Amherst College is tomorrow 12:10pm in McConnell B15!
- DataFest is a weekend-long “data science hackathon” for teams of up to 5 students and is at UMass the weekend of March 29-31! Info session by Prof. Randi Garcia today 7-8pm in Sabin-Reed 301 (here)!

- Recap of Lec05.
- Say I want visualize the distribution of temperature split by month. Two options
- Faceted histogram Figure 3.16
- Boxplot Figure 3.23

- Starting Data Wrangling: the pipe operator
`%>%`

and`filter()`

rows of a data frame.

moderndive readings in above schedule.

- Added note to syllabus on office hours:
*If you’re having R or RStudio issues, please have your computer and RStudio loaded and ready to go.* - ModernDive Chapters 4 & 5 are now reordered and renamed:
- Chapter 4:
~~Tidy Data via tidyr~~Data Wrangling - Chapter 5:
~~Data Wrangling via dplyr~~Data Importing and “Tidy” Data

- Chapter 4:
- Term project groups are due this Friday 5pm; see Term Project. If you need a group, Slack me.
- Lab tomorrow: Jenny will talk about DataCamp & cover data visualization.

- Recap of Lec04: Histogram binning structure & facets
- Boxplot to show the
*distribution*of a numerical variable split by a categorical variable. Say we want to plot a boxplot of the following 12 values which are pre-sorted:

1, 3, 5, 6, 7, 8, 9, 12, 13, 14, 15, 30

They have the following *summary statistics*:

Min. | 1st Qu. | Median | 3rd Qu. | Max. |
---|---|---|---|---|

1 | 5.5 | 8.5 | 13.5 | 30 |

- R Markdown:
- Don’t be afraid of error messages! In particular the line number where the error occurs!
- Heads up!
`View()`

nor`?`

will prevent your`.Rmd`

files from knitting (i.e. the HTML report won’t get created)!

- moderndive readings in above schedule.

- Slack message and
`#moderndive_typoes`

- Remember the
`Warning: Removed 5 rows containing missing values (geom_point).`

warning message you got when creating a scatterplot of`alaska_flights`

arrival and departure delays? Check out this talk by Prof. Brittney Bailey from Amherst College next Thursday 2/7 12:10pm in McConnell B15:

- Recap of Lec03.
- Histograms to show
*distribution*of a numerical variable.

- Live demo: Some tips on workflow for in-class exercises.
- “Typing/running code directly in console” vs “typing code in your
`class_notes.Rmd`

R Markdown file and*sending*it to the console to run.” - Quickly switching applications on your computer with “command + tab” (macOS) or “control + tab” (Windows)
- Making code human-friendly to read! Be empathetic to your collobators by writing nice code, in particular your most important collaborator!
- For example: hard returns between code chunks.
- See screencast of live demo below!

- “Typing/running code directly in console” vs “typing code in your
**Please close your RStudio Server window when not working! It uses up Smith server resources if you don’t!**- ModernDive readings for Lec04 in above schedule.

The BBC uses `ggplot2`

for data journalism!

The BBC releases its ggplot toolkit, bbplot! https://t.co/TMLI2kTMJn

— Kieran Healy (@kjhealy) February 1, 2019

- Slack message and
`#moderndive`

typos - Updated Term Project page with:
- Information on first phase: Form groups
- Example of final “resubmission” due the last day of class. Note this example is subject to change throughout the semester.

- Added all Term Project items to Moodle.

- Recap of Lec02:
`nycflights13`

package and`glimpse()`

/`View()`

functions. - What is a
*function*? What are*arguments*? - Grammar of graphics
- Scatterplots

ModernDive readings for Lec03 in above schedule. Now that you’ve seen R Markdown in Lab 1:

- Create a
`.Rmd`

file and save it as`class_notes.Rmd`

. That way you can save all your code for re-use later, like a Word document. - Copy and paste any code from ModernDive into “code chunks” in
`class_notes.Rmd`

. That way you can easily tweak/modify code. - “Run” code in the console from the “code chunks” in
`class_notes.Rmd`

as you learned in Lab 1. - Again, you do not need to submit any answers for learnings checks, however
**you are resposible for completing the readings, running all code, and doing all learning checks doing before the next lecture**.

- On Moodle
- If you haven’t already, please complete all the steps in “start here”
- If you are trying to register for this course, see the posted registration priority list.

- Outside help:
- Slack
- Did you get a Slack notification in some form for my message on Saturday at 9AM: mobile/desktop or email notification? You are responsible for staying on top of in-between lecture notifications.
- First student question posted on
`#questions`

Slack channel🎉 !!! The 🏆 for first student answer to a student question is still up for grabs!

- First lab with Dr. Jenny Smetzer is tomorrow. Problem set 01 (PS01) will be posted on Moodle.
~~the Problem Sets page~~.

Why chalk talks? Read the Field Notes slogan.

Why undirected in-class exercise time? People all learn at their own pace. What to do:

- Open RStudio (R in the menu bar above)
- Open ModernDive (ModernDive in menu bar above)
- As indicated in the above schedule, read ModernDive Chapter 2 while running all code in the
*console*. - You can skip all videos and the DataCamp links as we’ll be talking about those in class.
- You do not need to turn in Learning Checks, those are for your practice. The solutions are in Appendix D.
- If you have questions, ask a peer. If you’re still stuck, ask me!

Recall from the “How can I succeed in this class?” discussion in the syllabus:

Lectures, labs, and readings:

“Am I actually running the code and studying the outputs in R during in-class exercises, or am I just skimming the text?”“Am I completing all the ModernDive readings/in-class activites for a given lecture before the start of the next lecture?”“During in-class exercises and lab time, am I taking full advantage that I’m in the same place at the same time with the instructor, the lab assistants, and most importantly your peers, or am I browsing the web/texting the whole time?”Problem sets, DataCamp, and coding:

“When learning to code, much like learning a language, have I been really pushing myself to practice, practice, practice?”

- Please ensure you have followed all the “start here” instructions posted on Moodle. Please remember that
**just because you can access the moodle page does not guarantee you are registered for the course**. I will post a “priority list” of waitlisted students on Moodle by tomorrow. - What is the difference between SDS/MTH 220 vs SDS 201?
- Who is rudeboybert?
- Website features
- Slack demo
- Final project discussion
- Food for thought on coding:

New blog post!

— Olivia Guest | Ολίβια Γκεστ (@o_guest) November 26, 2018

👩🏻 💻👨🏿 💻👩🏾 💻👨🏽 💻👩🏼 💻👩🏿 💻👨🏻 💻

Why women in psychology can't program

“About two months ago my brother, who works in a data science on social psychology data, asked me why his colleagues, who are women and have PhDs in psychology, cannot code”https://t.co/5ZYx28sWr1