Which Technique For Representing Numeric Data Has Two Forms Of Zero?

Data is a valuable asset – so much so that information technology's the world's well-nigh valuable resource. That makes understanding the different types of data – and the role of a data scientist – more than of import than ever. In the business earth, more companies are trying to sympathize big numbers and what they tin exercise with them. Expertise in information is in high demand. Determining the correct data and measurement scales enables companies to organise, place, analyse and ultimately apply data to inform strategies that will allow them to make a genuine affect.

Data at the highest level: qualitative and quantitative

What is data? In curt, it'southward a collection of measurements or observations, divided into two different types: qualitative and quantitative.

Qualitative data refers to information most qualities, or information that cannot be measured. Information technology'south commonly descriptive and textual. Examples include someone's eye colour or the blazon of car they bulldoze. In surveys, it's often used to categorise 'yes' or 'no' answers.

Quantitative data is numerical. It'due south used to ascertain data that can be counted. Some examples of quantitative information include distance, speed, meridian, length and weight. It's easy to call back the divergence betwixt qualitative and quantitative information, as one refers to qualities, and the other refers to quantities.

A bookshelf, for example, may have 100 books on its shelves and be 100 centimetres tall. These are quantitative data points. The color of the bookshelf – red – is a qualitative data point.

What is quantitative (numerical) data?

Quantitative, or numerical, data tin be broken downwardly into two types: discrete and continuous.

Detached data

Detached data is a whole number that tin't be divided or broken into individual parts, fractions or decimals. Examples of discrete data include the number of pets someone has – one can take two dogs but not two-and-a-one-half dogs. The number of wins someone's favourite squad gets is too a form of discrete data considering a team can't have a one-half win – information technology'due south either a win, a loss, or a draw.

Continuous data

Continuous information describes values that can be broken down into different parts, units, fractions and decimals. Continuous data points, such equally height and weight, can exist measured. Time can too exist broken down – by half a second or one-half an 60 minutes. Temperature is another example of continuous data.

Discrete versus continuous

There's an easy way to remember the divergence between the two types of quantitative data: data is considered discrete if it can exist counted and is continuous if it can exist measured. Someone can count students, tickets purchased and books, while one measures height, distance and temperature.

What is qualitative (categorical) data?

Qualitative data describes the qualities of data points and is not-numerical. Information technology'south used to define the information and can also be farther broken downwardly into sub-categories through the iv scales of measurement.

types of data

Properties and scales of measurement

Scales of measurement is how variables are defined and categorised. Psychologist Stanley Stevens developed the four common scales of measurement: nominal, ordinal, interval and ratio. Each calibration of measurement has backdrop that determine how to properly analyse the data. The properties evaluated are identity, magnitude, equal intervals and a minimum value of goose egg.

Properties of Measurement

Identity: Identity refers to each value having a unique pregnant.
Magnitude: Magnitude means that the values have an ordered human relationship to 1 another, then at that place is a specific social club to the variables.
Equal intervals: Equal intervals hateful that data points along the calibration are equal, and then the deviation betwixt information points one and two will exist the same equally the deviation betwixt data points five and 6.
A minimum value of zero: A minimum value of zero ways the calibration has a true zippo point. Degrees, for example, can fall beneath zero and still accept meaning. But if you weigh nothing, you don't exist.

The 4 scales of measurement

By understanding the scale of the measurement of their information, data scientists can determine the kind of statistical examination to perform.

1. Nominal scale of measurement

The nominal scale of measurement defines the identity holding of data. This scale has certain characteristics, but doesn't have any form of numerical significant. The data tin can be placed into categories but tin can't be multiplied, divided, added or subtracted from one another. It'south also non possible to measure out the difference between data points.

Examples of nominal data include eye colour and country of nascence. Nominal data tin can be cleaved down again into iii categories:

Nominal with order: Some nominal data can be sub-categorised in order, such as "common cold, warm, hot and very hot."
Nominal without order: Nominal data can also exist sub-categorised as nominal without social club, such as male person and female.
Dichotomous: Dichotomous information is defined by having simply two categories or levels, such every bit "yes' and 'no'.

2. Ordinal scale of measurement

The ordinal scale defines data that is placed in a specific order. While each value is ranked, there'due south no information that specifies what differentiates the categories from each other. These values can't be added to or subtracted from.

An example of this kind of data would include satisfaction information points in a survey, where 'i = happy, two = neutral, and iii = unhappy.' Where someone finished in a race also describes ordinal data. While first place, 2nd place or third place shows what order the runners finished in, it doesn't specify how far the outset-identify finisher was in front of the second-place finisher.

3. Interval scale of measurement

The interval scale contains properties of nominal and ordered information, but the deviation between data points can be quantified. This type of data shows both the social club of the variables and the verbal differences between the variables. They can exist added to or subtracted from each other, but not multiplied or divided. For instance, twoscore degrees is not 20 degrees multiplied by two.

This scale is also characterised by the fact that the number nil is an existing variable. In the ordinal scale, zero means that the data does non exist. In the interval calibration, zero has meaning – for example, if you measure degrees, zero has a temperature.

Data points on the interval scale have the same difference betwixt them. The difference on the scale between ten and twenty degrees is the same between 20 and xxx degrees. This scale is used to quantify the deviation between variables, whereas the other two scales are used to describe qualitative values only. Other examples of interval scales include the year a car was made or the months of the year.

iv. Ratio scale of measurement

Ratio scales of measurement include properties from all four scales of measurement. The data is nominal and defined by an identity, tin can be classified in society, contains intervals and can be cleaved down into exact value. Weight, height and distance are all examples of ratio variables. Information in the ratio calibration can be added, subtracted, divided and multiplied.

Ratio scales also differ from interval scales in that the scale has a 'true zero'. The number zippo means that the data has no value point. An example of this is height or weight, as someone cannot be naught centimetres tall or weigh zero kilos – or be negative centimetres or negative kilos. Examples of the use of this scale are calculating shares or sales. Of all types of data on the scales of measurement, data scientists tin can practice the near with ratio information points.

To summarise, nominal scales are used to label or draw values. Ordinal scales are used to provide information well-nigh the specific order of the data points, by and large seen in the apply of satisfaction surveys. The interval calibration is used to sympathize the guild and differences between them. The ratio scales gives more than information about identity, guild and departure, plus a breakdown of the numerical detail within each information point.

Using quantitative and qualitative information in statistics

Once data scientists have a conclusive data set from their sample, they tin can start to use the information to draw descriptions and conclusions. To do this, they can employ both descriptive and inferential statistics.

Descriptive statistics

Descriptive statistics help demonstrate, stand for, analyse and summarise the findings contained in a sample. They present data in an easy-to-understand and presentable course, such as a tabular array or graph. Without description, the data would exist in its raw form with no caption.

Frequency counts

One way information scientists can describe statistics is using frequency counts, or frequency statistics, which describe the number of times a variable exists in a data set up. For example, the number of people with blue eyes or the number of people with a commuter's license in the sample tin can exist counted past frequency. Other examples include qualifications of pedagogy, such as high school diploma, a academy degree or doctorate, and categories of marital status, such as unmarried, married or divorced.

Frequency data is a form of discrete data, as parts of the values can't be broken downwards. To summate continuous data points, such as age, data scientists tin can use central trend statistics instead. To exercise this, they find the mean or average of the information signal. Using the age example, this can tell them the average historic period of participants in the sample.

While data scientists can draw summaries from the use of descriptive statistics and present them in an understandable form, they tin't necessarily draw conclusions. That's where inferential statistics come in.

Inferential statistics

Inferential statistics are used to develop a hypothesis from the data ready. It would be incommunicable to get data from an entire population, so information scientists can use inferential statistics to extrapolate their results. Using these statistics, they can make generalisations and predictions nigh a wider sample group, even if they oasis't surveyed them all.

An example of using inferential statistics is in an election. Even earlier the entire country has voted, data scientists can utilize these kinds of statistics to make assumptions regarding who might win based on a smaller sample size.

Using information visualisation to communicate insights

Information visualisation describes the techniques used to create a graphic representation of a data sample by encoding it with visual pieces of data. It helps to communicate the data to viewers in a articulate and efficient way.

Characteristics of constructive graphical displays

Effective visualisation can help individuals analyse complex data values and draw conclusions. The goal of this process is to communicate findings as clearly as possible. A graphic display that features effective messaging will show the information clearly and permit the viewer to gain insights and trends from the data set and reveal the different findings between the data.

Data visualisation examples

The best visual representation of a data prepare is adamant by the relationship information scientists want to convey between data points. Do they desire to present the distribution with outliers? Do they want to compare multiple variables or analyse a single variable over time? Are they presenting trends in your data ready? Here are some of the key examples of data visualisation.

A bar chart is used to compare two or more values in a category and how multiple pieces of data relate to each other.
A line nautical chart is used to visually represent trends, patterns and fluctuations in the data set up. Line charts are commonly used to forecast information.
A scatter plot is used to prove the relationship between information points in a compact visual grade.
A pie chart is used to compare the parts of a whole.
A funnel chart is used to represent how data moves through different steps or stages in a process.
A histogram is used to represent information over a sure time flow or interval.

Quantitative messages

Quantitative messages describe the relationships of the data. Depending on the sample, there are different ways to communicate quantitative data.

Nominal comparison: Sub-categories are individually compared in no detail society.
Time series: An private variable is tracked over a catamenia of fourth dimension, usually represented in a line nautical chart.
Ranking: Sub-categories are ranked in order, usually represented in a bar nautical chart.
Part-to-whole: Sub-categories are represented as a ratio in comparison with the whole, usually represented in a bar or pie chart.
Deviation: Sub-categories are compared with a reference point, unremarkably represented in a bar chart.
Frequency distribution: Sub-categories are counted in intervals, unremarkably represented in a histogram.
Correlation: Two sets of measures are compared to identify if they move in the same or opposite directions, usually represented in a besprinkle plot.

Expand your information science expertise

With data scientific discipline becoming a skill in fifty-fifty greater demand, now is a perfect time to expand your knowledge of the globe'south most valuable resource: data. A degree in data science volition enable you to place, analyse and present complex and interwoven webs of data. You tin can and so leverage these insights to make predictions and create strategies, specifically in a business environment. The UNSW Master of Data Science can requite you the skills you need to unlock the power of data and help businesses make better decisions, empowering them to drive meaning changes and results.