1. Data type and measurement scale
(1) Data type
There are two basic types of data: qualitative data and quantitative data.
Qualitative data : It is composed of non-numeric values. For example: color of shoes, gender.
Quantitative data : It is composed of values representing quantity or scale. For example: test scores.
(2) Discrete data and continuous data
- Discrete data : You can only take specific, individual values, not the middle of these values.
- Continuous data : can take any value in a given interval.
For example, the number of cows on the farm can only be rounded, so the data is discrete. The milk production of the cows on the farm can take any value within a certain range, so the data is continuous.
(3) Measurement scale
Another way to classify data is to classify it by measuring scale.
-
The data measured by classification is only the data consisting of name, code and category. (The data is qualitative and cannot be ranked and arranged)
-
Sequential measurement is suitable for qualitative data that can be arranged in a certain order (such as from high to low). It is usually meaningless to calculate the data measured by sequence.
-
Fixed-distance measurement is suitable for quantitative data where the spacing is meaningful but the ratio is meaningless. The data zero point is arbitrary.
-
Fixed ratio measurement is suitable for quantitative data with meaningful spacing and ratio. The data zero point is determined.
Second, error handling
(1) Error type:
- Random error : It is caused by random and inherently unpredictable events in the measurement process. Such as: weighing weight
- Systematic error : It is due to a problem with the measurement system. This problem has always affected all measurement results in the same way. Such as: errors caused by inaccurate timing due to damage to the timer.
(2) The size of the error: absolute error and relative error
- Absolute error : describes the difference between the measured value and the true value.
Absolute error = measured value (or claimed value)-true value - Relative error : It is to compare the size of the data error with the true value.
Relative error = absolute error / true value ✖ 100%
(3) Expression of results: accuracy and precision
Accuracy: Describes how close the measured value is to the true value.
Accuracy: Describes the level of detail in the measurement. For example, the scale used to weigh medicines is different from the scale used to weigh watermelons.
