Statistics Summary
Lectures and Practicum Notes
by Lidia Nikolova
BA CIS
2023-2024
,
, Block 1, week 1 - Lecture 1, 07/09/2023
Goals of this course:
- Get an understanding of descriptive (describing a population) and inferential statistics
(computing p-values and assessing if difference between numbers is significant)
- Statistical reasoning – by completing lecture exercises and practicums.
- There is practical approach, but some calculation are done to understand principles
- We will learn how to understand and apply basic analyses
- How to report statistical analyses and understand them
What we will learn in this course:
1. In terms of descriptive statistics we will learn:
The differences between measures for central tendency and spread.
Measure for central tendency is the mean (sum of all scores divided by the number of
participants)
Variation – by looking at the range (the minimum and the maximum value)
We can also summarize in terms of visualizations e.g., bar plot.
Descriptives is one variable at a time, summarizing that single variable for a sample.
2. Inferential statistics are about multiple variables (CIS and LING students) to generalize the
outcome of a sample to a population:
- Compare two groups (or a single group with fixed values like grades between males and
females)
- Check associations between two variables (e.g. is my English pronunciation related to the
grade I received in high school)
3. Internal consistency of questions in a questionnaire, how to run statistical tests in R
4. Creating lab reports
Basic Functions of R:
- Use as a calculator (adding, multiplying)
- Storing values in variables (a <- 5)
- Adjust a variable (b <- a*a, b= 25)
- d <- NA (if means is missing)
- Storing multiple values in a variable (b <- c (2, 4, 6, 7, 8) – function ‘c’ to
combine
- To add a number on a specific position b [4] <- a
▪ [1] 2 4 6 5 8
- Function ‘mean’ to calculate the mean score
- If we have missing values in our dataset the mean will not be computed, that’s why we
need a function that will ignore the NAs
▪ mean (b, na.rm = TRUE)
- To check the structure of the data frame use function str(), how many variables
, - To view parts of the data frame:
o Dat [a,b]
o a gives selected rows of dat – dat [1, ]
o b gives selected columns of dat – dat [, 1]
o don’t forget the comma
- data can be viewed by columns by means of names for the columns (variable names):
o dat[c(1 ,3, 5), c(“participant”, “study”) #rows 1, 3, 5 and 3
named columns
- A single column can also be selected by means of ‘$’ function: dat$gender (getting
the column gender of the dataset dat)
- Saving selected data tmp <- dat[5:8, c(1,3)]
Q1: How can we select the value in the third column, fourth row of dataset dat?
A. dat[4,3]
B. dat[3,4]
C. dat[3,]$4
D. dat[4,]$3
If within a data frame want to select specific cases, for example only interested in male
participants we can use CONDITIONAL INDEXING:
tmp <- dat[dat$gender == “M”,]
To combine conditions use ‘&’:
tmp <- dat[dat$gender == “M” & dat$study == “IS”]
To turn a conditional indexing around: ‘!=’
#only women (i.e., not men) *or* everybody with a grade higher than 7
Tmp <- dat[dat$gender != “M” | dat$english_grade > 7, ]
Add data to columns:
Dat$diff <- dat$english_grade – dat$english_score #this column will contain
info about the difference of English grade and English score
If I want to indicate whether people failed or passed then:
dat$pass_fail <- “pass” #new column, initially everybody passes
dat[dat$english_grade < 5.5, ]$pass_fail <- “fail” #if grade too low, then fail
head(datpdat$english_grade > 4 ^ dat$english_grade < 6,]) #show subset of data
If I create a new variable I first all the values to one specific value and then adjust the cases that
do not meet the criteria.
Basi options to visualize data in R;
- barplot() - barplot
- plot() - plot
