-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathfuture_changes
137 lines (109 loc) · 5.22 KB
/
future_changes
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
schedule
* order should be one-sample t-test, paired sample t-tests, independent-samples t, effect sizes (cohen's d and overlap), assumptions and non-parametric tests
general lab feedback
* make the first minihack easy and walk them through it
* find out what was in the bootcamp and you can skim or gloss that content
* exercises throughout are good
* try to budget at least 20 minutes for mini-hacks
* is it possbile to use the learnr package and have students download shiny apps to run on their computer?
* idea for the future: use learnr package to create labs, bundle these inside a package
* add in-text R code (reporting sample sizes, means, proper formatting, etc)
TA expectations
* prep labs (minor tweaks), attend lecture, take quizzes and homework assignments (to give feedback on difficulty and time)
homework general
* B's get degrees
* don't say "it's in your best interest to attempt all of them"
1-intro.html
* Replace Leibniz picture with Helmhotlz
* Don't bother with Locke and Leibniz
* The cleaner story is to talk about Descartes (statue story) then to Helmholtz (measuring reaction times) then more complex data.
* Hold back non-psych students to meet each other and form study groups
* Goals of this class:
1. learn stats and R
2. challenge you -- this class is designed to push you to your personal limit and then just beyond. if you feel like you're sailing through this class come talk to me
3. develop practical skills that will help throughout your career -- learning things on your own, asking for help, saying no, even teaching
* road map: summarizing data --> probability --> sampling --> making inferences about the unknown --> making a decision --> evaluating that decision
2-measurement.html
* Replace "units" with something else
* Remove humor nomological net -- better to use the student generated example
lab-1.html
* make it shorter -- didn't get to mini-hacks (aim for 20 min for mini-hacks each week)
* importing data can be delayed until lab 2 (maybe?)
* idea: keep all material in document, but skim sections (e.g., types of variables)
* next set of RAs add questions throughout (parts where students have to complete code or write answers)
* students look at prepared markdown, but TAs live-code (print html so you know where you are)
3-describing_data_1.html
* too long!
* did we use world happiness? take out description of variables
* leptokurtic - leaping up, platykurtic -- plate, mesokurtic -- middle
* is there a more efficient way to make the point about statistical tests of normality being tied to sample size?
lab-2.html
* more interactive throughout
* ask more questions (sometimes on html doc, or not)
* in-text R code summarizing sample size and means
5-matrix.html
* For XT and TX, give example in slides
* post-multiply -- affecting columns, doing this to every single row
* pre-multiply -- affecting row, doing this to every single column
* add orders to all matrix notations
quiz-2.docx
* replace word "masquerade"
lab-3.html
* notes: covered a lot of material, went slowly and left time for first mini-hack (walk through first together), linear combinations were difficult
* more relevant or straight-forward examples for diagonal matrices and pre- and post-multiplication
* give mean example
6-probability.html
* define random
* add animations!
* bag of candy -- replacement vs not
7-binomial.html
* make clear -- binary/binomial doesn't mean 50%
* any sample space can be divided into two non-elementary events
* explain a deck of cards (what?!)
1-homework.Rmd
* rewrite correlation freedom/generosity instructions: "calculate the correlation between freedom and generosity" (take out thing about "for a country")
* question 5.1 -- take out density distribution or give them code. (students are drawing based on mean and sd, not using empirical)
* 5.3 change validity of correlation to "statistical conclusion valdity"
* more time for homeworks? post 2 weeks in advance, instead of 1
8-normal.html
* move slides about chi sq from sampling to here
lab-4.html
* need to explain a deck of cards
10-nhst.html
* more on pvalues
* include figure with p-values bunching up at .05
* move ratio of signal to noise here
lab-5.html
* skim tidy data and helper functions, but keep in so people can see them
12-critiques.html
* fewer quotes at the beginning (be prepared to run out of time)
2-homework.html
* MIDDLE 50%
* clarifiy confidence interval 1-[alpha]
* summed SQUARED z-scores
* 20 pt question -- clarify that they should use norm, not t
lab-7.html
* is this worth having?
* next time, chi-square
* doing the test
* frequency tables
* output tables to word document
* extracting information
* in-line R code to summarize results
* or more on data wrangling
* move summarize here
* more about first one
* tidy data
* pros: first year sem already covers some of this?
14-one-sample.html
* reiterate signal/noise
15-two-means_I.html
* reiterate signal/noise
lab-8.html
* include welch's t-test
* good to have them walk through the steps
* had time to go through the mini-hacks
17-dependent means.html
* the cohen's d discussion is awkward but healthy. keep it in
* pay attention to notation for standard deviation and standard error
* no matter which d you use, say which one or indicate how you calculated