University of Chester

School of Computer Science, Mathematics and Business Computing

MSc Information Systems

 

Research Methods 1,
Professional Standards and Issues

CO7101

 

 

Task 3

 

A Local Education Authority wishes to introduce an electronic method for registering pupils in classes. The education officers believe that if teachers use a handheld electronic device which is linked to the school’s database over a wireless network, then office staff will learn of pupils deliberately missing specific lessons in a school day much sooner, allowing staff to be deployed more quickly to find the pupils.

 

Two makes of handheld device are considered for purchase for the teachers. The LEA agrees with the sellers of the devices that it will trial both devices for four weeks each, throughout 4 schools. You are brought in as a consultant to analyse the relative success of each device, so that a final decision can be made by the LEA.

 

You must decide how the four schools are chosen (you can assume you will have enough devices for any size of school), what data will be collected over the four weeks and how it will be analysed. You must also think about how you will present the results to the LEA.

 

In about 500 words or equivalent, describe briefly how you will undertake this task. It is recommended that you make up a small amount of example data to demonstrate how you will use SPSS to test your data, Include printouts of this demonstration of SPSS in your assignment.

 

Marks will be awarded for addressing the following issues:

 

1.                  How to choose the 4 schools

2.                  The data you will collect, including the type

3.                  Choosing appropriate data collection techniques

4.                  Choosing an appropriate method of analysis

5.                  Choosing appropriate methods of data presentation, whether it is for your benefit or to be shown to the LEA.

 

 

Please email your answers directly to Graham Logan on g.logan@chester.ac.uk no later than 20th December 2012


Selection of Schools

This is an ideal opportunity to make a direct immediate impact. The selection of schools to take part in the LEA’s pilot is a simple one: we want the worst of the worst. Our target schools are all secondary and have the highest rates of truancy, by current recording standards, the worst league table performances and average pupil numbers of 1,800. The schools all share the following common characteristics: weak or no monitoring of daily attendance, inconsistent attendance policies, lack of parent involvement in the school, lack of personalized attention to students, lack of teacher expectations for high student achievement.

Data to be collected

Variable Name

Type

Width

Values

StudentID

Numeric

10

Unique

Date of birth

Date

 

DD/MM/YYYY

Sex

Numeric

1

0 = Male

1 = Female

Ethnicity

String

3

IC1 White person

IC2 Mediterranean person

IC3 African/Caribbean person

IC4 Indian, Nepalese, Pakistani, Maldivian, Sri Lankan, Bangladeshi, or any other (South) Asian person

IC5 Chinese, Japanese, or South-East Asian person

IC6 Arabic, Egyptian or Maghreb person

IC7 Mixed race person

IC0 Origin unknown

School

Numeric

1

0 = school 1

1 = school 2

2 = school 3

3 = school4

Key stage

Numeric

1

0 = key stage 3

1 = key stage 4

Year

Numeric

1

 

SubjectID

Numeric

2

0 = Religious education

1 = ICT

2 = English

3 = Citizenship

4 = Art and Design

5 = Design and Technology

6 = Geography

7 = History

8 = Mathematics

9 = Modern foreign language

10 = Music

11 = PE

12 = PSHEE

13 = Physics

14 = Chemistry

15 = Biology

Date

Date

 

 

StartTime

Time

 

 

EndTime

Time

 

 

Product

Numeric

1

0 = product 1

1 = product 2

attendance

Numeric

1

0 = present

1 = absent with permission

2 = absent

DevicePreference

Numeric

1

0 = product 1

1 = product 2

 

 

Data collection

Data collection will be automatically digitally captured. Each student will be given, randomly, one of the two trial devices at the start of each lesson. The student will logon to the device, confirming identity, school and lesson. The data to be collected is aggregated into attendance lists and distributed to the class room teacher, head of subject, head of year, head master, truancy intervention team, LEA.

 

A follow up survey of all participating pupils will also be digitally captured using an online questionnaire.

Analysis

The sample data will comprise:

analysis will comprise:

the data will be tested for normality

comparison of truancy rates between male and female

comparison of truancy rates across year groups

comparison of truancy rates across ethnic classification

comparison of truancy rates between schools

Presentation

 

Tests of Normality

 

Sex

Kolmogorov-Smirnova

 

Statistic

df

Sig.

attendance

0

.223

216000

.000

1

.223

216000

.000

a. Lilliefors Significance Correction

 

The data is normally distributed


comparison of truancy rates between male and female

 

Correlations

 

Sex

attendance

Sex

Pearson Correlation

1

-.001

Sig. (2-tailed)

 

.486

Sum of Squares and Cross-products

108000.000

-187.000

Covariance

.250

.000

N

432000

432000

attendance

Pearson Correlation

-.001

1

Sig. (2-tailed)

.486

 

Sum of Squares and Cross-products

-187.000

287701.374

Covariance

.000

.666

N

432000

432000

 

 

comparison of truancy rates across year groups

 

Correlations

 

attendance

Year

attendance

Pearson Correlation

1

.001

Sig. (2-tailed)

 

.714

Sum of Squares and Cross-products

287701.374

278.000

Covariance

.666

.001

N

432000

432000

Year

Pearson Correlation

.001

1

Sig. (2-tailed)

.714

 

Sum of Squares and Cross-products

278.000

864000.000

Covariance

.001

2.000

N

432000

432000

 


comparison of truancy rates across ethnic classification

 

As is evidenced there is an absolutely even distribution between rates of truancy and ethnicity

 

comparison of truancy rates between schools

 

Correlations

 

attendance

School

attendance

Pearson Correlation

1

.001

Sig. (2-tailed)

 

.602

Sum of Squares and Cross-products

287701.374

313.000

Covariance

.666

.001

N

432000

432000

School

Pearson Correlation

.001

1

Sig. (2-tailed)

.602

 

Sum of Squares and Cross-products

313.000

540000.000

Covariance

.001

1.250

N

432000

432000

 

 

The data is a bust! Because I used random number generators for some of the values it generates a linear even distribution of all the key values.

 

What I had intended to show was that attendance had improved throughout the trial, that product 2 was the preferred device by the students and that there were many other interesting things to say.