Stats Tutorial 1 Notes: Intro to
stats
Statistics VS Parameter
Statistics (Sample)
- Representative of a characteristic of a sample of people (a smaller
group from a large population)
- Keywords include: a percentage/number of a certain population
- Example of a statistic value: “60 male shoppers at Aventura
Mall were randomly chosen and their waistlines were measured
revealing an average waistline of 37 inches” 60 male shoppers is
only a sample of shoppers at Aventura Mall and not a representation
of all the male shoppers in at Aventura Mall
Parameter (Population)
- Representative of a characteristic of an entire population
- Keywords include all/every the (something/group) in that area
- Example of a parameter value: “A study of every national park’s
sales receipt for August shows they brought in an average of 10
million dollars of revenue” Every national park’s sales receipt can
be a representation of the average revenue that is earned by
National parks in that country
Sample VS Population
Sample
- It includes only a portion, of part, of the population of interest
- Example: Small groups of students from ACC, ITB and BS are
selected to do a survey in NP it is only a portion of students from
the whole of BA (a population)
Population
- It is a collection of all possible individuals, objects or measurements
of interest
- Example: All students from School of BA are selected to do a
survey about BA all students in BA are included in the collection
of information for the survey about BA
Quantitative VS Qualitative
Quantitative
- Refers to data that involves numbers and can be measured
- Example: height of person, number of apples, speed of a car
There are 2 types of quantitative data
1. Discrete Data
- They are countable numbers, meaning they are whole numbers
(integers) with no fractions or decimals
- Example: Number of apples, O-level scores, number of cars parking
in a carpark
2. Continuous Data
- They are measurable numbers that can have any value, including
decimals or fractions
- They can take on an infinite number of values within a given range
- Example: A person’s weight, average O-level scores, time to finish a
race
Qualitative
- Refers to non-numerical data that describes qualities or
categories
- However, there are exceptions where it involves numbers but is still
considered as qualitative data as the numbers mean a label or
characteristic of something and not necessarily presents itself as a
data or statistic. (Example: Postal Code of an address, model of a
phone/computer)
- It is about the characteristics or labels of something
- Example: colours, types of music, opinions
Levels of Measurement
Nominal Level
- Most basic level of measurement
- Data are categorised into groups or labels, but there is no order or
ranking
What can be measured: Only the categories or groups the data fall into,
you can count how many items fall into each category
What is not meaningful: There is no natural order or ranking in the
categories. Example, you can’t say that one category (e.g. eye colour) is
better or more than another
Summary: You can label or name the categories, but no ranking or
meaningful comparison between the categories exists
Examples of Nominal Level Data: Eye colour, gender, religious affiliation
Ordinal Level
- This level involves data that can be ordered or ranked, but the
differences between the rankings are not measurable or meaningful
What can be measured: You can rank the data in order of magnitude (e.g.
1st, 2nd or 3rd), there is an inherent sense of order
What is not meaningful: The difference between the ranks is not
consistent or measurable, you don’t know how much more one item is
compared to another, you cannot say that the difference between the
ranks is equal
Summary: You can rank the data, but differences between the ranks
don’t have a consistent or measurable meaning
Examples of Ordinal Level Data: Satisfaction level found in surveys
Explanation of Example:
- This data is ordered because "Very satisfied" is better than
"Satisfied," and "Very dissatisfied" is worse than "Dissatisfied." So,
there's a clear ranking of responses.
- However, the differences between these categories aren't
measurable. You can't say exactly how much more satisfied
someone is who chose "Very satisfied" compared to someone who
chose "Satisfied." The gap between "Satisfied" and "Neutral" might
not be the same as between "Neutral" and "Dissatisfied."
Interval Level
- Here, the data can be ordered, and you can measure the exact
difference between values. However, there is no "true zero" point,
meaning zero does not represent the absence of what is being
measured, and it's just another point on the scale, not a true zero
point.
- Example: Temperature in Celsius or Fahrenheit. The difference
between 20°C and 30°C is the same as between 30°C and 40°C, but
0°C doesn’t mean "no temperature."
What can be measured: You can measure the difference between values,
and they are meaningful. The intervals (gaps between values) are
consistent
What is not meaningful: There is no true zero or baseline. Zero does not
mean “nothing”. Because of this, ratios are not meaningful as you can’t
say one value is “twice as much” as another. (e.g. the difference between
20°C and 40°C is a 20-degree difference. However, 0°C doesn't mean "no
temperature," so you can't say 40°C is "twice as hot" as 20°C.)
Summary: You can measure consistent differences between values, but
there is no true zero, so ratios (multiplying/dividing values) are
meaningless.
Examples of Interval Level Data: Temperature, shoe sizes
Ratio Level
- This is the highest level of measurement. Data at this level can be
ordered, differences can be measured, and there's a meaningful
zero point, which means zero represents the absence of the
quantity.
- Example: Height, weight, age, or income. Zero dollars means no
money, and you can compare values in a meaningful way (someone
earning $200 earns twice as much as someone earning $100).
What can be measured: You can measure consistent differences
between values, and there is a true zero, meaning that zero represents
the absence of the quantity being measured. You can also compare
values in terms of ratios you can say one value is "twice as much" as
another.
What is not meaningful: Nothing is not meaningful here this is the
most complete level of measurement.
Summary: You can measure differences, there is a true zero, and you can
compare values using ratios (multiplying or dividing). Everything is
meaningful here.
Examples of Ratio Level Data: Monthly income of surgeons, or distance
traveled by sports school students per month
