Comprehensive Guide to Data Handling and Statistics

Fundamentals of Data and Variable Types

Data consists of individual factual pieces of information that are recorded and utilized for analytical purposes. To make sense of data, it must be sorted and organized. Data can be obtained from any source depending on the specific content being searched for.

There are five primary types of data categorizations:

1. Bivariate Data: This type of data involves two variables. It is used to investigate both sets of data to identify if there are any connections between them. An example is investigating the marks achieved in Maths and English to see how they relate.

2. Univariate Data: This involves a single variable or one type of data. It describes a specific attribute or characteristic. An example includes the height of students in a class, recorded as values such as $158\,cm$, $160\,cm$, and $149\,cm$.

3. Discrete Data: This data can be counted and contains a limited number of values. It is represented in whole numbers or integers and cannot be broken down into smaller fractional parts. Examples include the number of learners in a classroom or the number of people in a swimming pool.

4. Continuous Data: This type of data is measured rather than counted. It can take on an unlimited number of different values because it is not fixed, and it often includes decimal values. Examples include temperature, weight, or the time taken to finish a task.

5. Categorical Data: This is a form of information that is stored and identified based on names or labels. It is used to group data into specific categories. Examples include gender (girl, boy), housing type (house, plat, shack), eye colors, favorite subjects, or types of pets.

Data Collection and Sampling Methodologies

When collecting data, two main groups are considered:

1. Sample: This is a small part of a quantity intended to show what the whole population is like. It represents a larger population and is used because it is smaller and easier to manage.

2. Population: This refers to the entire group about which conclusions are intended to be drawn.

Sampling can be categorized into biased or unbiased methods:

• Biased Sample: This occurs when the number of subjects being sampled for a specific attribute is not equivalent or proportional to the number of subjects that effectively have that attribute within the population.

• Random or Unbiased Sample: This occurs when every member of the population has an equal chance of being included in the sample. A random sample must accurately represent the population.

Guidelines for sample size relative to population size are as follows:

• Small population (less than $500$ people): The sample size should be $20\%\text{ to }30\%$.

• Medium population ($500\text{ to }1000$ people): The sample size should be $10\%\text{ to }15\%$.

• Large population (more than $1000$ people): The sample size should be $5\%\text{ to }10\%$.

Measures of Central Tendency

A central tendency is a single value that attempts to describe a data set by identifying the central position within that data. The three common measures are:

1. Mode: This is the value that occurs most often in a data set. A mode does not exist if there are an equal number of all items. Example: in the set $1, 2, 3, 3, 3, 3, 4, 4, 5, 5, 6, 7, 8, 8$, the mode is $3$.

2. Median: This is the middle value of a data set when the values are sorted or arranged in ascending order. It can only be used for numbers and measurements. One drawback is that it can take a long time to put large datasets in order. If there are two middle numbers, such as in the set $1, 2, 3, 4, 5, 6$, the median is calculated as $\frac{3+4}{2} = 3.5$.

3. Mean: This is the sum of all data in the set divided by the number of data points. A few very large or very small numbers can significantly affect the size of the mean. This measure can only be used for numbers and measurements. Example: for the set $1, 2, 3, 4, 5, 6, 7, 8, 9, 10$, the mean is the sum divided by $10$.

Quartiles and Interquartile Range

Quartiles consist of three values ($Q_1$, $Q_2$, and $Q_3$) that divide a data set into four approximately equal parts, with each part containing $25\%$ of the elements. To split quartiles evenly, the data must be univariate (a single variable) and must be ordinal, interval, or ratio data. The specific positional points are:

• Minimum: The lowest value at the $0^{th}$ percentile.
• $Q_1$ (Lower Quartile): The $26^{th}$ percentile.
• $Q_2$ (Middle Quartile/Median): The $50^{th}$ percentile.
• $Q_3$ (Upper Quartile): The $75^{th}$ percentile.
• Maximum: The highest value at the $100^{th}$ percentile.

To calculate quartiles when there are two middle numbers, they must be added together and divided by $2$. For example, $\frac{4+5}{2} = 4.5$ for $Q_1$.

The Interquartile Range (IQR) is the distance between $Q_1$ and $Q_3$. It is calculated as: $IQR = Q_3 - Q_1$

Outliers can be determined using the following formulas:

• Lower outliers: $Q_1 - (1.5 \times IQR)$
• Higher outliers: $Q_3 + (1.5 \times IQR)$

Example calculation from a data set:

• Minimum: $3$
• $Q_1$: $6.5$
• $Q_2$ (Median): $8.5$
• $Q_3$: $12.5$
• Maximum: $16$
• $IQR$: $12.5 - 6.5 = 6$
• Lower outlier threshold: $6.5 - (1.5 \times 6) = -2.5$
• Higher outlier threshold: $12.5 + (1.5 \times 6) = 21.5$
• Conclusion: There are no outliers in this specific set.

Data Distribution Shape and Skewness

The shape of the distribution describes how data is spread across the set:

• Symmetric Distribution: Both sides of the box plot or distribution are mirror images of each other, where the median ($Q_2$) is perfectly centered.

• Positively Skewed Distribution: The tail on the right side of the distribution is longer than the left side.

• Negatively Skewed Distribution: The tail on the left side of the distribution is longer than the right side.

Graphical Representation of Data

Various formats are used to represent data visually:

• Bar Graphs: Use rectangular bars where the length of the bar is proportional to the value represented. Key features include gaps between categories, bars of equal width, and proper labeling of the axes. An example is a graph showing favorite fruit (Apple, Orange, Pears, Banana) where the vertical axis is the number of fruit.

• Histograms: Used for continuous numerical data. Unlike bar graphs, the bars touch each other to represent numerical intervals or ranges, such as height categories ($145-150\,cm$, $151-155\,cm$).

• Pie Charts: A circular statistical graphic divided into slices to illustrate numerical proportion. The whole graph represents $100\%$. Slices represent portions of the whole.

  • To calculate the degree of a slice: $\frac{\text{Value}}{\text{Total}} \times 360^{\circ}$. Example: $\frac{12}{36} \times 360^{\circ} = 120^{\circ}$.
  • To calculate the percentage of a slice: $\frac{\text{Value}}{\text{Total}} \times 100$. Example: $\frac{24}{96} \times 100 = 25\%$.

• Pictographs/Pictograms: These charts use icons or pictures to represent data. They are specifically easier for children to understand. A key must be included to define what each icon represents. For example, one smiley face might represent $4$ children.

Importance of Data Visuals in Cognitive Development

Utilizing graphs and data representations is important for several educational areas:

• Number Sense: Developing a deep understanding of numbers and the number system, including the ability to compose and decompose numbers.
• Algebra: Understanding patterns and relationships, which includes the processes of sorting and categorizing.
• Measurement: The ability to make comparisons, order items, and understand measurable attributes.
• Data Analysis & Probability: Understanding data as a means of conveying organized information and realizing how certain variables affect that data.

Scatter Plots and Correlation Analysis

A scatter plot is a graph that helps determine if there is a correlation between two sets of numeric data. It uses two axes, one for each variable.

• $x$-coordinate represents the independent variable.
• $y$-coordinate represents the dependent variable.
• Each ordered pair of values is plotted as a single point.

Example application: A scatter plot can be used to see if there is a relation between age and grade. For instance, data points like ($13, 16$), ($16, 6$), and ($14, 12$) suggest older learners are often in higher grades.

Correlation types include:

• Zero Correlation: Points are scattered randomly, indicating no pattern between the two sets of data and no correlation between variables.
• Strong Positive Correlation: Points form a band that slopes upward from the bottom left to the top right. As one variable increases, the other also increases.
• Strong Negative Correlation: Points form a band that slopes downward from the top left to the bottom right. As one variable decreases, the other increases/decreases accordingly.
• Moderately Positive Correlation: Points are clustered from bottom left to top right, but not as closely as in a strong positive correlation.
• Moderately Negative Correlation: Points are clustered from top left to bottom right, but not as closely as in a strong negative correlation.

Other relationships include:

• Exponential Relationship: Points follow a curve from left to right, showing exponential growth or decay.
• Quadratic Relationship: Points move in a characteristic "up then down" or "down then up" parabolic shape.

When a linear association exists between two variables, it is called a two-variable correlation. A strong linear correlation allows for the drawing of a "line of best fit," also known as a regression line or the least squares regression line.

Standard Deviation

Standard deviation is a measure of how spread out the values in a data set are from the mean average.

• Small Standard Deviation: Values are close to the mean, and the data has less variation. Example data set: $8, 9, 10, 11, 12$.
• Large Standard Deviation: Values are further from the mean, and the data has more variation. Example data set: $2, 6, 10, 14, 18$.
• Standard Deviation of $0$: This occurs when all the values in a data set are exactly the same.

Data Handling in the Foundation Phase

In the primary education stage, data handling follows a specific progression:

1. Collect and sort physical objects: Sorting items like shapes, stationery, or objects by size, color, or shape.
2. Represent sorted collection of objects: Drawing pictures of the collected items.
3. Discuss and report on sorted collections: Answering questions about how the sorting was performed, how the collection looks, and describing the drawing of the collection.
4. Collect and organize data.
5. Represent data.
6. Analyze and interpret data.