Lecture 2 Variables & Measurement PDF

AEDF 0623 VARIABLES & MEASUREMENT Introduction to Statistics for Built Environment CONTENTS 1) Definition of variable and data 2) Statistical data source 3) Types of data/variables 4) Levels/scales of measurement of variables 5) Univariate, bivariate & multivariate 6) Independent...

AEDF 0623 VARIABLES & MEASUREMENT Introduction to Statistics for Built Environment CONTENTS 1) Definition of variable and data 2) Statistical data source 3) Types of data/variables 4) Levels/scales of measurement of variables 5) Univariate, bivariate & multivariate 6) Independent and dependent variables  DEFINITION OF VARIABLE 1 Any measured characteristics; any attribute, amount, or number that is countable or measurable A variable is a characteristic or property of the 2 units being studied, for value might fluctuate over time and between data units in a population E.g., age, gender, business revenue and costs, place 3 of birth, capital expenditure, class grades, type of vehicle, and types of drainage system.  DEFINITION OF DATA A set of values or observations, which are 1 collected through reliable sources to achieve the goals of study. 2 Often arranged according to observational units (people in the sample/population) and variables (what is being measured). 3 The quality of data gathered determines the quality of the study.  STATISTICAL DATA SOURCE SOURCE OF DATA Direct Method Indirect Method PRIMARY DATA SECONDARY DATA Collecting information from every Data are collected as part of the day- unit in the population (a census), or to-day processes and record from a subset of units (a sample) keeping of organisations i.e., data from the population made available by others Surveys, interviews/FGD, Historical data or public records i.e., observational studies, experiments books, journal, reports, etc  TYPES OF DATA &VARIABLE TYPES OF DATA/VARIABLE QUALITATIVE QUANTITATIVE (Categorical/Attribute) (Numerical) Data that refers only to name Data that represent counts or classifications measurements Continuous Discrete (measured (counted items) State, types of vegetation, level of characteristics) education, satisfaction level Temperature, time, No. of rooms, defects IQ test, CGPA, per hour, number of weight, height, days distance  QUALITATIVE DATA 1 Labels or names used to identify an attribute or characteristic of each element. 2 Represented in the form of separate distinct categories. 3 Example: o Type of vegetation: groundcover, shrubs, trees o LAM Part 3 Exam: Pass or fail o Residents’ race: Malay, Chinese, Indian  QUANTITATIVE DATA 1 Always numeric and indicate either “how many” or “how much” 2 In the form of measurements and counts, can be ranked and ordered. 3 Example: o Distance o Height o Weight o Speed  QUALITATIVE & QUANTITATIVE Q u a l i t at i v e variable Q u a l i t at i v e D at a Q u a nt i t at i v e variable Q u a nt i t at i v e D at a  QUALITATIVE & QUANTITATIVE Q u a l i t at i v e variable Q u a l i t at i v e D at a Retrieved from:https://www.mdpi.com/1660 -4601/19/16/10017  QUALITATIVE & QUANTITATIVE Q u a l i t at i v e variable Q u a l i t at i v e D at a  QUALITATIVE & QUANTITATIVE Q u a l i t at i v e variable Q u a l i t at i v e D at a  QUALITATIVE & QUANTITATIVE Q u a nt i t at i v e variable Q u a nt i t at i v e D at a Retrieved from: https://mrkremerscience.com/2013/08/28/making - Q u a nt i t at i v e D at a scientific-data-tables/ QUALITATIVE & QUANTITATIVE Q u a nt i t at i v e variable Retrieved from: https://www.researchgate.net/publi cation/264976319_A_CRITICAL_AN A LY S I S _ O F _ R O A D _ A C C I D E N T S _ V E H I CULAR_FLOW- Q u a nt i t at i v e D at a _ R AT E _ A N D _ S P E E D _ W I T H _ R E F E R E N CE_TO_ROUTE_F050_BP-KLUANG QUALITATIVE & QUANTITATIVE Q u a nt i t at i v e variable Retrieved from:https://www.researchgate.n et/publication/283159577_Reco mmended_Chair_and_Work_Surfa c e s _ D i m e n s i o n s _ o f _ V D T _ Ta s k s _ f o r_Malaysian_Citizens/figures?lo= 1 Q u a nt i t at i v e D at a  QUALITATIVE & QUANTITATIVE Q u a l i t at i v e Q u a nt i t at i v e variable variable Q u a l i t at i v e D at a Retrieved from: https://www.mef.org.my/news/general_article.aspx?@ID=59 Q u a nt i t at i v e D at a DISCRETE & CONTINUOUS Quantitative variables can be classified as; Discrete Continuous DISCRETE & CONTINUOUS DISCRETE o A variable with possible scores of discrete points on the scale o Why discrete? – Because discrete variables can only be integers or whole numbers o Example: Number of defects per hour, number of cubicle in a toilet, number of brick blocks CONTINUOUS o The scale is continuous and NOT made up of discrete steps o Measurements can take on any value/any degree of precision o Example: height, weight, distance, temperature DISCRETE & CONTINUOUS Number of lightbulb installed? DISCRETE DISCRETE & CONTINUOUS Speed of wind? CONTINUOUS DISCRETE & CONTINUOUS Height of wave? CONTINUOUS DISCRETE & CONTINUOUS Consumed electricity in kilowatt CONTINUOUS DISCRETE & CONTINUOUS Number of color palette? DISCRETE DISCRETE & CONTINUOUS Distance from and to a place? CONTINUOUS DISCRETE & CONTINUOUS Number of cars? DISCRETE LEVEL/SCALES OF MEASUREMENT OF VARIABLES Expressions that typically refer to the theory of scale types developed by the psychologist Stanley Smith Stevens All measurement in science was conducted using four different types of scales: LEVEL/SCALES OF MEASUREMENT OF VARIABLES NOMINAL 1 3 INTERVAL ORDINAL 2 4 RATIO LEVEL/SCALES OF MEASUREMENT OF VARIABLES Ratio HIGHEST LEVEL Complete Analysis. Measurement (weight, height, temperature, income) Interval HIGHER LEVEL Ordinal Mid-level analysis Ranking/ordered categories (level of education, medal) Nominal LOWEST LEVEL Basic analysis Identification, name, categories LEVEL/SCALES OF MEASUREMENT OF VARIABLES Nominal o Consists of assigning items to groups, names, labels, categories or ID o No quantitative information is conveyed o No ordering of the items is implied o Used to measure qualitative variables only o Example: Types of trees, race, types of wall finishes, paint color NOMINAL SCALE LEVEL/SCALES OF MEASUREMENT OF VARIABLES Ordinal o Ordered in the sense that higher numbers represent higher values o The intervals between the numbers are not necessarily equal and the differences are meaningless o Allow us to rank order the items in terms of “which has less?” and “which has more?” o CANNOT say “how much more?” o Example: Education level, economic status, Likert-scale, satisfaction level ORDINAL SCALE LEVEL/SCALES OF MEASUREMENT OF VARIABLES Interval o A scale in which a certain distance along the scale means the same thing no matter where on the scale you are, but where "0" on the scale DOES NOT represent the absence of the thing being measured o Allows us not only to rank order the items that are measured, but also to quantify and compare the sizes of differences between them o Differences are meaningful o Example: Temperature, IQ test INTERVAL SCALE LEVEL/SCALES OF MEASUREMENT OF VARIABLES Ratio o Very similar to interval variables o It features an identifiable absolute zero "0" point o 0" on the scale represent the absence o Example: Weight, height, distance RATIO SCALE Preferred number of rooms among house buyers in KL LEVEL/SCALES OF MEASUREMENT OF VARIABLES Mark out of Mark relative to Student Position Result 100% 40% pass mark R AT I O I N T E R VA L ORDINAL NOMINAL Ahmad 56 16 6 Pass Ali 48 8 7 Pass Cassey 65 25 3 Pass Daud 73 33 2 Pass Elias 62 22 4 Pass Fatima 35 -5 10 Fail Sayyed 20 -20 9 Fail Hana 38 -2 8 Fail Nurul 58 18 5 Pass Zaleha 82 42 1 Pass Game ranking ORDINAL Book color NOMINAL Duration of presentations RATIO Time (1-2pm) INTERVAL Quantity of coffee in a cup RATIO Smartphone brand NOMINAL Yearly temperature INTERVAL  UNIVARIATE, BIVARIATE & MULTIVARIATE DESCRIPTION & ANALYSIS UNIVARIATE BIVARIATE MULTIVARIATE More analytical than Complex form of statistical Data contains only one univariate analysis analysis technique variable Data set contains two More than two variables in Does not deal with a causes variables the data set or effect relationships To compare between the To understand the To describe the data to find two data set relationship of each variable patterns within the data Measure the correlations with each other. through data presentation between the two Can be analysed using Mean, median, mode, variables. Principal Component Analysis dispersion, variance, range, Can be analysed using or logistic regression, linear standard deviation etc. correlation, scatter plot regression, cluster analysis  UNIVARIATE, BIVARIATE & MULTIVARIATE EXAMPLES UNIVARIATE BIVARIATE MULTIVARIATE Example: Based on the season, we cannot predict the weather of any Example: given year. Several factors play The example of bivariate Example: an important role in predicting can be surrounding The example of a univariate can the same. temperature and how long be height of trees. Multivariate analysis offers a the air conditioning system more complete examination of is being switched on. data by looking at all possible independent variables and their relationships to one another INDEPENDENT VARIABLE DEPENDENT VARIABLE VS INDEPENDENT VARIABLE o Manipulated by the experimenter o The subjects/factors Example: A study on the effectiveness of four different types of antidepressants medicine. DEPENDENT VARIABLE Independent variable (being o Measured from the subjects manipulated) –type of antidepressant used o Dependent upon the subjects/factors Dependent variable – relief from depression  VS INDEPENDENT VARIABLE DEPENDENT VARIABLE Discuss: A study on the effectiveness of classroom design and students’ performance. 1) Which one is the IV and DV? 2) How would you want to measure your variables? 3) What type of data would you like to obtain? 4) Can this study also be performed with multivariate analysis?  VS INDEPENDENT VARIABLE DEPENDENT VARIABLE Discuss: A study on students' satisfaction towards WIFI connection in Mahallah. 1) Which one is the IV and DV? 2) How would you want to measure your variables? 3) What type of data would you like to obtain? 4) Can this study also be performed with multivariate analysis?  VS INDEPENDENT VARIABLE DEPENDENT VARIABLE Discuss: A study on the amount of water towards plants’ height after several days. 1) Which one is the IV and DV? 2) How would you want to measure your variables? 3) What type of data that you would like to obtain? 4) Can this study also be performed with multivariate analysis? Thank You Wassalam