In my previous post, we discussed some basic statistical concepts I learnt during my first data analytics session at LBS; they include Qualitative and Quantitative data, Population and sample size. Today, we will discuss other terminologies such as Cross-sectional data and Time series data, Confidence Level, Significance level, Outliers, Parameter, Skewness, Degree of freedom, One or Two-tailed tests.
Cross-sectional data are data that are observed, collected, and recorded simultaneously, providing a snapshot of the data at a given time.
Time series data are data that are collected over a duration of time. This data can be collected annually, quarterly, or biannually. Also, note that a combination of cross-sectional and time series data is the Pull or Panel data.
The Confidence interval or level is the percentage of the correctness of your analysis. It communicates how accurate our estimate is likely to be.
Level of significance is the room for error in data analysis. It is important to note that your margin of error should be minimal when the research being carried out is very sensitive. The significance level for Best practice is 5%.
An Outlier is an observation that is abnormally different from the other observations. In other words, they are unusual values in a dataset. Outliers are important because they can change the results of our data analysis.
A Parameter is a numerical summary of a population. We can also say it is the single value used to summarize or describe a population.
Causality is when the cost and effect are connected or when an effect is traced back to its cause. For example, if the price of goods increases and causes sales to drop, there is a causality.
Skewness is the measure of the Degree of departure from symmetry (normal distribution) which depends on the comparative positions of the mean, median and mode. The graph of a positively skewed distribution moves toward the right, while a negatively skewed distribution moves toward the left.
One or Two-tailed test – A one-tailed test is used to ascertain if there is any relationship between variables in a single direction, either left or right. In contrast, the two-tail test is used to identify whether or not there is any relationship between variables in either direction. To simplify this, calculating both the positive and negative sides of a graph is a two-tailed test, while one side is a one-tailed test.
The Degree of freedom means adjusting for a sample size. It is a combination of how much data you have and how many parameters you need to estimate. To get the Degree of freedom, you take your sample size and subtract one. For example, n – 1, where n is the sample size.
Measurements (NOIR) simply means how variables are measured, or we can say it is the scale or levels of measurement assigned to a variable. They are Nominal, Ordinal, Interval, and Ratio.
- The Nominal level of scale is a naming scale, where variables are simply “named” or labelled, with no specific order. In Norminal scale, numbers, such as driver’s license numbers, are used to name or identify people, objects or events.
- The Ordinal scale is mostly used to measure qualitative data. Ordinal data can be counted and ordered but not measured. For example, the ranking of school students – 1st, 2nd, 3rd, etc.
- Interval level data is ordered like the Ordinal data, but the intervals between each value are known and equal. For example, the Fahrenheit and Celsius temperature scales are examples of interval measurements.
- Ratio level of scale is the highest level possible for data. Ratio data is ordered with known and measurable intervals between each value. The physical characteristics of persons and objects can be measured with the ratio scale; thus, height and weight are good examples of ratio measurements. A person weighing 45kg is two-thirds as heavy as a person who weighs 68kg.
#MEMBA11
Written by Tutty Tero
