Assignment/ Project Using SPSS

Here, the sample size, n=81

Exercise 1

a) Compute the Mean, Median, Quartile, 65th Percentile, Decile, Standard Deviation, and
Variance & Range of Selling Prices

Multiple modes exist. The smallest value is shown

Interpretation:
 Mean: $225.49 is the typical selling price of the homes.
 Median: 50% of the prices of homes are below $213.70 and 50% are above it.
 65th Percentile: 65% of the prices of homes are below $239.80 and 35% are above it.
 Standard Deviation: The Standard Deviation of the selling price of homes is $46.470
 Variance: The Variance of the selling price of homes is 2159.444 (Dollar Square).
 Range: The Range of the selling price of homes is $219.

Quartile:
Q1= 25% of price of homes are below $191.65 (thousand) and 75% are above it. Q2= 50%
of price of homes are below $213.70 (thousand) and 50% are above it. Q3= 75% of price of
homes are below $255.85 (thousand) and 25% are above it.

Decile:
D1= 10% of price of homes are below $176.88 (thousand) and 90% are above it. D2= 20% of
price of homes are below $ 184.36 (thousand) and 80% are above it. D3= 30% of price of
homes are below $191.65 (thousand) and 70% are above it. D4= 40% of price of homes are
below $208.04 (thousand) and 60% are above it. D5= 50% of price of homes are below $
213.70 (thousand) and 50% are above it. D6= 60% of price of homes are below $ 228.16
(thousand) and 40% are above it. D7= 70% of price of homes are below $ 249.64 (thousand)
and 30% are above it. D8= 80% of price of homes are below $267.16 (thousand) and 20%
are above it. D9= 90% of price of homes are below $295.88 (thousand) and 10% are above it.

Percentile:
P65th= 65% of price of homes are below $ 239.80 (thousand) and 35% are above it.
 b) Develop a Bar Diagram and also a Pie chart for the variable “Township”

Figure: Bar Diagram

Interpretation:

Uttara and Banani has the highest percentage of township which is 27.16% & Nikatan has the
lowest percentage of township which is 13.58%. Dhanmondi has the second highest percentage
of township which is 17.28%. Gulshan has a 14.81% of township which is closer to the
parcentage of township of Nikatan.
 Figure: pie chart

Interpretation:

Uttara and Banani has the highest percentage of township which is 27.16% & Nikatan has the
lowest percentage of township which is 13.58%. Dhanmondi has the second highest percentage
of township which is 17.28%. Gulshan has a 14.81% of township which is closer to the
parcentage of township of Nikatan.
Exercise 2
a) Determine the co-efficient of skewness for the variable “selling price”. Is the distribution
positively or negatively skewed

Findings: The co-efficient of skewness for the variable “selling price” is .592

Interpretation: The distribution is Weakly Positively Skewed

Exercise 3

a) Develop a 99 percent confidence interval for the mean selling price of the homes.

Findings: One-Sample Statistics

Interpretation:
The Mean selling price ranges from $211.86 - $239.11 and we are 99% confident about it.
b) Develop a 95% percent confidence interval for the mean distance of the home from the centre
of the city.

Interpretation:

The mean distance of the home from the centre of the city ranges from 13.4593 Km – 15.729 Km
and we are 95% confident about it.

Exercise 4
a) A recent article in The Bangladesh Observer indicated that the mean selling price of the
homes in the city of Dhaka is more than Tk.220 (Thousand). Can we conclude that the
mean selling price in the Gulshan area is more than Tk.220 (Thousand)? Use the 0.05
significance level.

Step 1: State the Null & Alternate Hypothesis

Ho: µ ≤ 220

H1: µ > 220

Step 2: Select the Level of Significance

Step 3: Determine the test statistics. From the table, t- Test statistics is used.

Step 4: Formulate the decision rule:

If the P-value is larger than the level of significance, H0 will be accepted.

Otherwise rejected.

Step 5: Here, P/2 = 0.291/2 = 0.1455

 Since P value (0.291) > α (0.05). H0 Is not rejected. So µ ≤ 220.

Step 6: Interpretation: No, we cannot conclude that the mean selling price in the Gulshan
area is more than $220 (thousand) at the .05 significance level.

b) The same article reported that the mean size was more than 210 square feet. Can we
conclude that the mean size of homes sold in the Gulshan area is more than 210 square
feet? Use the 0.05 significance level.

Step 1: State the Null & Alternate Hypothesis

Ho: µ ≤ 210

H1: µ > 210

Step 2: Select the Level of Significance

∞ = .05

Step 3: Determine the test statistics.

From the table, t- Test statistics is used.

Step 4: Formulate the decision rule.

If the P-value is larger than the level of significance, H0 will be
accepted.Otherwise rejected.

Step 5: Here, Since P value (0.000) < α (0.05). H0 is rejected. So µ >210.

 Step 6: Interpretation: Yes, we can conclude that the mean size of homes sold in the
Gulshan area is more than 210 square feet at the .05 significance level.

Exercise 5

(a) I. Simple linear regression equation

We know, the regression equation is y´=a+bx

From the Coefficients table, a= 85.222 (Y intercept), b= 0.063 (regression coefficient) Here, y
´= price of the homes (dependent variable), x= size of homes in sq.ft (independent variable),

So our regression equation, y´=85.222 +0. 063 x

Interpretation: 1. a=85.222 is the point through which the regression line crosses the Y axis. 2.
If size of the home (x) is increased by 1 sq.ft, the price of the home (y´) is increased by
0.063dollar.

From the Model Summery table,

Coefficient of correlation, r = 0.333

Interpretation: There is a weak and positive relationship between size of homes in sq.ft (x) and
Price of the homes (y´).

Coefficient of determination, R2 = 0.111

Interpretation: 11.1% of the total variation in the dependent variable Price of the homes (y´) is
explained or accounted by the variation in independent variable size of homes in sq.ft (x).

Std. Error of the Estimate, Sy.x = 44.099

Interpretation: 44.099 is the typical error we make when we use the regression equation to
estimate the dependent variable (Price of the homes y´).
5 (a) II

We know, the regression equation is y´=a+bx

From the Coefficients table, a= 164.450 (Y intercept), b= 15.597 (regression coefficient)

Here, y´= price of the homes (dependent variable), x= number of bed rooms (independent
variable),

So our regression equation, y´=164.450 +15.597 x

Interpretation: 1. a=164.450 is the point through which the regression line crosses the Y axis.
2. If number of bed rooms (x) is increased by 1, the price of the home (y´) is increased by
15.597 dollar.
From the Model Summery table,

Coefficient of correlation, r = 0.333

Interpretation: There is a modarate and positive relationship between number of bed rooms (x)
and Price of the homes (y´).

Coefficient of determination, R2 = 0.111

Interpretation: 11.1% of the total variation in the dependent variable Price of the homes (y´) is
explained or accounted by the variation in independent variable number of bed rooms (x).

Std. Error of the Estimate, Sy.x = 44.099

Interpretation: 44.099 is the typical error we make when we use the regression equation to
estimate the dependent variable (Price of the homes y´).

5. (a) III

We know, the regression equation is y’=a+bx

From the Coefficients table,
a=121.464 (Y intercept),
b1=10.372, b2=.031, b3= -2.893, b4=17.108 (regression coefficient)
Here, y´= Price of the homes (dependent variable),
x1= Size of bed rooms (independent variable),
x2= Number of homes(independent variable),
x3= Distance from the city center (independent variable),
x4= Number of bathrooms (independent variable).
So our regression equation, y´=121.464 + 10.372x1 +.031 x2 – 2.893 x3 + 17.108 x4

Interpretation:
a=121.464 is the point through which the regression line crosses the Y axis.
If number of bed rooms (x1) is increased by 1, the price of the home (y´) is increased by 10.372
dollar.
If size of the home (x2) is increased by 1 sq.ft, the price of the home (y´) is increased by
0.031dollar.
If distance from the city center (x4) is increased by 1 km, the price of the home (y´) is decreased
by 2.893 dollar.

If number of bathroom (x3) is increased by 1, the price of the home (y´) is increased by
17.108dollar.

(b)

From the Model Summery table,

Coefficient of determination, R2 = 0.419
Interpretation:
38% of the total variation in the Price of the homes (y´) (dependent variable) is explained or
accounted by the variation in the number of bathrooms (x1), Size of homes (x2), distance from
the city center (x3), number of bed rooms (x40(independent variables).

(c). From the Model Summery table, Std. Error of the Estimate, Sy.x = 36.333

Interpretation:
36.333 is the typical error we make when we use the regression equation to estimate the
dependent variable (Price of the homes y´).