sandr063

1.       Understanding the Data v What  are  variables? v What  are  the  constants?                 v (Chances  of constants are very less but there can be; variables are required v actually) 2.      To Determine the Level of Measurement v What level is there; nominal/ordinal/interval or ratio. 3.      To Try to Find out Variables of Interest 4.      Distributions

Three Methods of Identifying Skewness 1. By looking at histogram and normal curve 2. By looking at the value of skewness (skewness must be zero for normal distribution, negative for negatively skewed and positive for positively skewed) 3. Relationship between mean, mode and median.

 KURTOSIS PEAKEDNESS OF A DISTRIBUTION LEPTOKURTIC:  high and thin MESOKURTIC:  normal in shape PLATYKURTIC:  flat and spread out   •   KURTOSIS measures how peaked the histogram is •   The KURTOSIS of a NORMAL DISTRIBUTION is 0 •   KURTOSIS characterizes the relative PEAKEDNESS or FLATNESS of a distribution compared to the normal distribution.  PLATYKURTIC When the kurtosis < 0, the frequencies throughout the curve are closer to be equal (i.e., the curve is more flat and wide). Thus, negative kurtosis indicates a relatively flat distribution. LEPTOKURTIC When the kurtosis > 0, there are high frequencies in only a small part of the curve (i.e, the curve is more peaked). Thus, positive kurtosis indicates a relatively peaked distribution  •   KURTOSIS is based on the size of a distribution's tails. •   NEGATIVE kurtosis (PLATYKURTIC) - distributions with short tails •   POSITIVE kurtosis (LEPTOKURTIC)...

 SKEWNESS Skewness measures the degree of asymmetry exhibited by the data If SKEWNESS equals zero, the histogram is SYMMETRIC about the mean -  Absence of symmetry -  Extreme values in one side of a distribution.

Levels of Measurement 1. Classification 2. Order (class position; 1st, 2nd, 3rd etc.) 3. Equal Intervals (exactly equal; inch, cm kg etc. reproduce able) 4. True Zero    

Degree of Freedom: Out of four values three (n − 1) have freedom; we can change their value to get the same mean. Mean is insensitive for individual value −−−− Along with this measure of dispersion is used. Range:    Inter quartile range Range:    Semi-Inter quartile range Range:    Standard Deviation

WHAT ARE THE MEASURE OF CENTRAL TENDENCY? Mean, Mode and Median Levels and central tendencies are very much related. If the level of measurement is nominal then central tendency will be only mode. For ordinal level median will be used. Level : What is the mean of 10, 20, 30 and 40? Answer: 10 + 20 + 30 + 40 = 100 Mean = 100 ÷ 4 =25 Median: The central value of data is called median; For example a set of values is  10,12,14,16,18,20,22 the central vale 16 is median. Note: If a set may be consist on even numbers like as 2 4 6 8 10 12 the central two value collect and divided by two For example the above set 6, 8 are two central vales 6+8 = 14 ..... 14/2 = 7 The number 7 is Median. Mode: The most frequent value is called MODE. In a set the frequent vale is always mode for example a set of data is consist 2,4,4,6,7,8,9 the most frequent value 4 is mode. NOTE: Some time two combine vale may be mode in this situation collect both values and divided by two like as 2, 4, 4, 6, 8, 8, 10 I...

FOUR LEVELS OF MEASUREMENT OF SOCIAL REALITIES 1.   NOMINAL LEVEL OF MEASUREMENT: Classification (on the base of some attribute) Example; male & female etc. We just clssify. 2.   ORDINAL LEVEL OF MEASUREMENT: Not only classify but order it too. Ordering; classification & ordering 3.   INTERVAL LEVEL OF MEASUREMENT: Yardstick; divided into equal intervals. Equal intervals are very important (Note that 1st, 2nd & 3rd are not intervals but the order) 4.   RATIO LEVEL OF MEASUREMENT: True zero concept