Everyone gave it five stars
A page says: "Customers rate our tool 5 out of 5." You look at the small print. Two people answered.
Nothing on the page is false. But the page left something out. Today you learn three signs that a source leans.
Key words
| Word | What it means |
|---|---|
| Bias | an unfair lean towards one side or one idea |
| Partiality | showing only one side and treating the other side unfairly |
| Prejudice | a judgement about a group of people made without fair evidence |
| Omission | leaving out a fact that the reader needs |
| Loaded word | a word chosen to make you feel something, such as amazing or terrible |
| Sample size | how many people or records a result is based on |
| Indicator | a sign that something may be true |
The cells students fill in are left blank in this preview.
By the end of this lesson
Note
I can find partiality, prejudice and omission in a source, and explain why each is a sign of bias.
Starter
Have a go
Starter
3 questionsYour answers are marked here and are not kept. A college account keeps every student’s marks.
Three signs of bias inside a source
| Sign | What it looks like | Data-work example |
|---|---|---|
| Partiality | One side only, with loaded words | "This amazing tool is the only sensible choice." |
| Prejudice | A judgement about a group with no fair evidence | "Older customers never use apps, so ignore their data." |
| Omission | A needed fact is left out | "Rated 5 out of 5" and no sample size |
The cells students fill in are left blank in this preview.
Partiality and loaded words
A fair source gives both sides. A partial one does not. Look for loaded words such as amazing, worst or only. They push your feelings and carry no fact.
Count them. A page with many loaded words and few figures is leaning.
Run it: count the loaded words
words = ["this", "amazing", "tool", "is", "the", "only", "sensible", "choice"]
loaded = ["amazing", "only", "best", "worst", "terrible"]
count = 0
for word in words:
if word in loaded:
count = count + 1
print("Loaded words:", count)Omission: what is missing
Omission is hard to see because the missing fact is not on the page. Ask: how many people answered? Over what time? What was left out of the chart?
A fair source gives the sample size next to every result.
Run it: the number hides the sample
select count(*) as responses, avg(rating) as average_rating
from survey;