LEVELS OF MEASUREMENT

Levels of Measurement, Measures of Central Tendency and Dispersion



Levels of Measurement, Measures of Central Tendency and Dispersion

I: Levels of Measurement, Measures of Central Tendency and Dispersion

From the ESS identify

one variable measured at the nominal level

These are numeric variables whose values ??represent a category or identify a group of belonging. This type of variable only allows establishing relations of equality/inequality between the elements of the variable. An example of such variables is the Gender and we can assign a value to the men and women a different and more sexist or feminist that we could not establish that one is greater than the other.

Statistics

Gender

N

Valid

42903

Missing

97

Mean

1.54

Median

2.00

Mode

2

Std. Deviation

.498

Range

1

Percentiles

25

1.00

50

2.00

75

2.00

one variable measured at the ordinal level

These are numeric variables whose values ??represent a category or identify a set of membership counting in a logical order. This type of variables allows us to establish relations of equality / inequality and in turn we can identify if a category is greater than or less than another. An example of ordinal variable is the level of education, since it can establish that a person has a graduate degree level of education higher than a person with a bachelor's degree.

Statistics

Highest level of education

N

Valid

41812

Missing

1188

Mean

3.01

Median

3.00

Mode

3

Std. Deviation

1.476

Percentiles

25

2.00

50

3.00

75

4.00

one variable measured at the ratio level

These are variables whose values ??represent numerical magnitudes and the distance between the numbers of the scale is the same. With this type of variables we can make comparisons of equality / inequality, establish order in their values ??and measure the distance between each value of the scale. For example the design weight is calculated under the ratio level measurement.

Statistics

Design weight

N

Valid

43000

Missing

0

Mean

.999981

Median

1.000000

Mode

1.0000

Std. Deviation

.4217815

Percentiles

25

.828600

50

1.000000

75

1.078500E0

The value of mean is 0.999 which indicates that most of the respondents are male.

one dichotomous variable

These are the dummy variable. In dummy variable a qualitative quantity is define by assigning a specific number. In the given ESS data file countries and TV watching, total time on average weekday are dummy variable.

Statistics

TV watching, total time on average weekday

N

Valid

42859

Missing

141

Mean

4.20

Median

4.00

Mode

7

Std. Deviation

2.028

Percentiles

25

3.00

50

4.00

75

6.00

The value of mean is 4.20 which indicates that most of the respondents watching TV for more than 1.5 hours, up to 2 hours.

Choose one variable measured at the nominal, ordinal, and ratio level from the ESS: calculate appropriate measures of central tendency for all three.

Statistics

Number of people living regularly as member of household

N

Valid

42936

Missing

64

Mean

2.76

Median

2.00

Mode

2

Std. Deviation

1.435

The value of mean is 2.76 which indicate that 3 people living regularly as member of household.

Choosing an appropriate variable in the ESS, calculate its interquartile range and explain the result.

Statistics

Number of people living regularly as member of household

N

Valid

42936

Missing

64

Percentiles

25

2.00

50

2.00

75

4.00

From the above table the value of quartiles can be observed. So the value of interquartile range is 2.

Choosing an appropriate variable from the ESS, calculate its mean and standard deviation and explain the result.

Statistics

Number of people living regularly as member of household

N

Valid

42936

Missing

64

Mean

2.76

Std. Deviation

1.435

The value of mean is 2.76 and standard deviation is 1.435 so it can be said that a little bit variation is present in the data set of number of people living regularly as member of ...