Course code BFT 2074 Course Title BIOMETRY AND

Course code: BFT 2074 Course Title BIOMETRY AND EXPERIMENTAL DESIGN Observed data & their Characteristics Prof Dr Md Ruhul Amin

Introduction and Data Collection 1. 1 Some definitions q Statistics: Statistics is a study dealing with the process of collecting, collecting organizing, organizing summarizing, summarizing analyzing and presenting (COSAP) information. q Population: Population is the totality of items or things under consideration possessing certain characteristics of interest. q Parameter: Parameter (yardstick) is a summary measure that describes a characteristic of an entire population. q Sample: Sample is the representative portion of the population that is selected for analysis. q A statistic is a summary measure computed from sample data that is used to describe or estimate a characteristic of the entire population.

…. Definitions Descriptive statistics Inferential statistics Descriptive statistics is the method that focus on the collection, presentation and characterization of a set of data in order to properly describe the various features of that set. Inferential statistics is the method of estimating the characteristics of a population or the e. g. Mean height of SBS students: 5’ making of a decision concerning a population based only on sample results. e. g. This one is better than that one

Definitions… Variable: Variable a variable is any measured characteristic or attribute that differs for different subjects. For example, if the weight of 30 subjects were measured, then weight would be a variable. If no. of students in different classes were counted then no. of students counted would be a variable.

Biometry �Statistics applied in the field of Life Science is called BIOMETRY or BIOSTATISTICS Life Science includes Biological Science, Medical Science, Agricultural Science

Why data are needed? �Provide the necessary input to a survey �Provide the necessary input to a study �Measure the performance of an ongoing service or production process �Evaluate the conformance of standards �Assist in formulating alternative courses of action in the decision making process �Satisfy our curiosity

Observation of a particular event Generally an observation can be classified as either QUANTITATIVE or QUALITATIVE. Quantitative observations are based on some sort of measurement or count eg. Length, weight, temperature and p. H, number of balls in the basket. Qualitative observations are based on categories reflecting a quality or characteristics of the observed event eg. Male vs female, diseased vs healthy, live vs dead, coloured vs colourless etc. Any observation when recorded is called DATA.

Types of variable 1. Quantitative variable • a. Continuous variable • b. Discrete variable 2. Ranked or ordinal variable • Example: Voters classified by parties • Students classified according to height 3. Categorical or qualitative variable • Examples Male vs Female • Red vs White

Variables or Data types There are several data types that arise in statistics. Each statistical test requires that the data analyzed be of a specific type. Most common types of variables 1. Quantitative variables – fall into two major categories a) Continuous variables- can assume any value in some (possibly unbounded) interval of real numbers. Common examples include length, weight, temperature, volume and height. They arise from MEASUREMENTS. b) Discrete variables- assume only isolated values. Examples include clutch size, trees per hectare, teats per sow, no. of days per month, no. of patient for a particular disease in hospitals. They arise from COUNTING.

Variables or Data types… 2. Ranked data (ordinal variables) are not measured but nonetheless have a natural ordering. For example, candidates for political affiliation can be ranked by individual voters. Or students can be arranged by height from shortest to tallest and correspondingly ranked without being measured. A candidate ranked 2 is not twice as preferable as the person ranked 1. 3. Categorical data or qualitative data: Some examples are species, gender (M/F), healthy vs diseased and marital status (married/ unmarried). Unlike ranked data, there is no ‘natural’ ordering that can be assigned to these categories.

1. Examples of data types Data type Question type Responses Numerical How many balls are in the basket ? Number How tall you are? Categorical ……. Inches/cm 1. Do you have any work experience? Yes or No 2. Name the types of victims in street accidents Killed or injured or unaffected

2. Example of nominal scaling Categorical variables Categories Colour of ball in the basket Blue / Red /Yellow/ Black Marital status Single / Married /Widow

3. Example of ordinal scaling Categorical variables Ordered categories Students grades ABCDEF Product satisfaction Unsatisfied Neutral Victims of street accident Died / Seriously injured / Slightly injured / Intact Satisfied

4. Example of interval and ratio scaling Numerical value Level of measurement Temperature Interval STANDARD Exam Score Interval Height, weight, age, salary Ratio

Collecting data �Primary data - the data that are gathered by researcher or data collector �Secondary data (source data) are the data obtained from data reservoir/data bank Once you have decided to use either secondary data or primary data or both, the next step is on how to collect the data. To collect secondary data is not a big problem. Just to approach the authority. Primary data collection needs specific design to have accurate and representative data at a minimum cost and time.

Reason for drawing a sample 1. A sample is less time consuming 2. A sample is less costly to administer than a census 3. A sample is less cumbersome and more practical to administer than a census Note: A sample must be representative for specific population/subpopulation

Table and graphs � The data collected in a sample are often organized into a table or graph as a summary representation. The following table shows the no. of sedge plants found in 800 sample quadrats (1 m 2 ) in an ecological study of grasses. Example 1. A frequency distribution table Table 1. Plant/quadrat Frequencies (fi) Total (xi ) 0 268 1 316 2 135 3 61 4 15 5 3 6 1 7 1 800

Example 2. The following data were collected by randomly sampling a large population of rainbow trout. The variability of interest is weight (lb) Xi (lb) fi fi X I 1 2 2 2 1 2 3 4 12 4 7 28 5 13 65 Total 27 109

Example 2…. �Rainbow trout have weights that can range from almost 0 -20 lb or more. Moreover their wt. s can take any value in that interval. For example, a particular trout may weigh 4. 3541 lb. From example 2 � lb.

A sample of bar graph 14 12 10 8 Series 3 Series 2 6 Series 1 4 2 0 Category 1 Category 2 Category 3 Category 4

A sample of bar graph…. �Category �Series Categories may be: 4 different states in Malaysia Series may be people 1. Bumi putra 2. Chinese origin 3. Indian origin

Pie chart A 15% B D 28% 50% C 7% A B C D

Line diagram 80 Daily expenditure (RM) of a week 70 60 50 40 30 20 10 0 Sun Mon Tues Wednes Thurs Fri Satur

Example of a chart Month 2011 Travel abroad JAN x FEB Exam Plantati on X X MAR APR X X Confere nce In Kl X X MAY X JUNE X X JULY X AUG SEP X X x X OCT X NOV DEC In home X x X X

Exercises 1. For each of the following random variable determine whether the variable is categorical or numerical. If numerical, determine whether the variable of interest is discrete or continuous.

Exercise 1 No. of telephones per household. Type of telephone primarily used. No. of long-distance call made per month. Length (minute) of long-distance call made per month. Colour of telephone primarily used. Monthly charge (RM) for long-distance call made. No. of local call made per month. Whethere is a telephone line connected to a computer modem in the household.