Question:

Assignment Task: data-analysis assignment

This assignment is in two parts: analysis of text (qualitative data analysis) and analysis of numbers (quantitative data analysis). Students must complete both parts. Each part carries equal weight.

Background/Context/Detail:

Part 1: Text analysis

The focus of this part is on demonstrating the technical skills of analysing text using thematic analysis techniques. We recommend the “framework” approach (Ritchie and Spencer, 1994) which has been introduced in class. You can combine it with “template analysis” (King 2004) which is similar in many respects.

The skills we asking you to demonstrate are those of:

(1) identifying those parts of the text relevant to answering your research questions

(2) identifying or creating a set of categories into which to classify the data

(3) creating an annotated list of the categories (the code book)

(4) categorizing segments of data, labelling them with the category names or labels (‘codes’), and validating the consistency of this activity

(5) creating data matrices or tables to ‘extract’ and assemble data relevant to each research question (or sub-question)

(6) using the data matrices to explore the data and to identify patterns in the data

(7) producing a short report answering your research questions, thus demonstrating your ability to begin writing up the analysis

analysing a set of interview transcripts

You will be provided with transcripts of four interviews with construction project managers about their work. The transcripts have been anonymised to protect the identity of the company, interviewees and interviewers. In these transcripts “education” and “power” refer to types of construction project.

The aim of the analysis is to understand what construction project managers’ work is like from their perspective. The research questions that you need to answer are:

1) What is a project manager’s job like in terms of

– Who do they report to, and who reports to them?

– Who are their ‘stakeholders’?

– What tasks and activities do they perform?

2) What do they like least and most about their work?

3) Why are they doing this job?

The most efficient way to accomplish the tasks is to load the texts into Excel at the earliest opportunity, and work in Excel throughout the analysis stages. If you do any preliminary work on hard copies of the transcripts, you may need to copy this into Excel.

(We strongly advise you against attempting to use any dedicated qualitative data analysis software like Nvivo. It takes a long time to learn how to use these tools effectively, it is not necessary, and it is best to practise the techniques without using these tools.)

Part 2: Number analysis

The data set was drawn from the British Household Panel Survey (BHPS). You do not need to know anything else about this data set, but if you are interested you can find further information here: https://www.iser.essex.ac.uk/bhps. The research questions for this section are:

1. a) Is there a difference in overall job satisfaction between workers from different geographical locations (i.e., London vs. rest of South East Britain)?

1. b) What happens to this potential difference in overall job satisfaction when additionally accounting for workers’ satisfaction with work content and workers’ satisfaction with pay?

For question 1. b), you need to consider the effects of geographical location, workers’ satisfaction with work content, and workers’ satisfaction with pay in one analysis.

2. a) Is there an interrelation between number of children in household and working hours per week?

2. b) Does this potential interrelation between number of children in household and working hours per week differ for female and male workers?

For question 2. b), try to find a way to account for number of children in household, gender, and the interaction of both variables (i.e., ‘number of children in household’ by ‘gender’) in one analysis.

The sample is quite large. Accordingly, it is likely that you find significant differences and interrelations. Therefore, you need to carefully evaluate the size of differences and the strength of interrelations.

Quantitative data-analysis tasks

Start with a description of the study sample in terms of demographics and work-related variables using descriptive statistics. After that, address the research questions above.

For each of the research questions, you have to

• Select an appropriate inferential statistical technique to answer the research question.

• Do the analysis.

• Report the following: (i) State your objective, (ii) explain why you chose which inferential statistical technique, (iii) present and explain the statistical result, (iv) present appropriate descriptive statistics and graphical illustrations that help to understand the result, (v) describe the finding in words (i.e., what does the statistical finding actually mean), and (vi) interpret the finding (i.e., how could the finding be explained?).

You need to present outputs of your inferential statistical analyses in the report (i.e., insert SPSS outputs in the text). You should pay special attention to descriptive statistics, to graphical displays, and to the interpretation of findings. Please note that interpreting findings means more than describing findings in words.

Reflective log/diary

You should keep and submit a reflective log/diary of the assignment work.

Answer:

# Part 1: Text Analysis

# Answer 1

## Part i

Interviewee 10103b is not very clear as to whom he reports to. His line manager is “remote.”

Interviewee 10104b’s work is to provide options to his clients and take directions from his client.

Interviewee 10105b reports to the to the programme director “a lady called Bernadette,” who is a consultant for Croydon.

For interviewee 10106b he is not very “clear” as to whom he reports to. Although he has “remote – line manger.”

## Part ii

Interviewee 10103b and 10104b work for “The company.” Interviewee 10105b and 10106b work for “Croydon.”

## Part iii

In Interview 10103b it is seen that the interviewee is in charge of preparation of documentation of the buildings. The documentation would be required by the organization for the functioning of the building.

In Interview 10104b and 10105b the interviewees work as project managers, overseeing the delivery of the work. Moreover interviewee 10105b works as accounts director also for the project.

Interviewee 10106b’s work is to oversee the housing project. There are 5 contract within the project and he is looking after the housing project.

# Answer 2

Interviewee 10103b and like the “variety” in their job. The respondents believed that their job offered enough of diversity in their jobs. In addition they were fond of the opportunity to “do different things everyday”

Interviewee 10103b said that he liked “freedom” in his work.

For interviewee 10105b the “professional drive”, “success” and his work which “make a difference” were very important.

Interviewee 10104b said that he “had limited autonomy” in his job.

# Answer 3

Interviewee 10103b is in his present profession since he studied course related to “procuring building and how they are put together.”

Interviewee 10104b is doing his present job since he has been in the profession of “costing”, “design” “planning” and then “actual development.”

Interviewee 10105b has been involved in “construction projects.”

Interviewee 10106b has been a “freelance surveyor.”

# Part 2: Number Analysis

## Part 1a

The objective of the present analysis to test whether there is difference in the level of job satisfaction between residents of London and rest of South East Britain.

The hypothesis for the test:

Null Hypothesis: There is no difference between the average level of job satisfaction between residents of London and rest of South East Britain.

Alternate Hypothesis: There is difference between the average level of job satisfaction between residents of London and rest of South East Britain.

To test the objective independent sample t-test is used. The independent sample t-test is used since:

- The dependent variable level of job satisfaction is defined in a continuous scale.
- There are two independent variables – the residents of geographical locations – Britain and rest of South-East Britain.

Table 1: Group Statistics |
||

overall job satisfaction (one-item measure) | ||

geographical location | ||

London | Rest of South East Britain | |

N | 158 | 185 |

Mean | 5.51 | 4.91 |

Std. Deviation | .982 | 1.226 |

Std. Error Mean | .078 | .090 |

Table 2: Independent Samples Test |
||||

overall job satisfaction (one-item measure) | ||||

Equal variances assumed | Equal variances not assumed | |||

Levene’s Test for Equality of Variances | F | 3.096 | ||

Sig. | .079 | |||

t-test for Equality of Means | t | 4.936 | 5.022 | |

df | 341 | 339.655 | ||

Sig. (2-tailed) | .000 | .000 | ||

Mean Difference | .599 | .599 | ||

Std. Error Difference | .121 | .119 | ||

95% Confidence Interval of the Difference | Lower | .360 | .364 | |

Upper | .838 | .834 |

The average (SD) level of job satisfaction of people of London is 5.51 (0.981) and of rest of South-East Britain is 4.91 (1.226).

The test statistics for the test is t(341) = 4.936. At 0.05 level of significance the p-value < 0.001. Since p-value < 0.05, level of significance, hence we reject the Null Hypothesis. Thus we can say that there statistically significant differences in the level of job satisfaction between the residents of London and rest of South-East Britain.

Hence it can be inferred that the average level of job satisfaction of the residents of London (5.51±0.991) is much higher than the residents of rest of South-East Britain (4.91±1.225).

## Part 1b

The objective of the present analysis to test whether there is difference in the level of work satisfaction with work content between residents of London and rest of South East Britain.

The hypothesis for the test:

Null Hypothesis: There is no difference between the average level of work satisfaction with work content between residents of London and rest of South East Britain.

Alternate Hypothesis: There is difference between the average level of work satisfaction with work content between residents of London and rest of South East Britain.

Table 3: Group Statistics |
||||

satisfaction with work content (one-item measure) | satisfaction with pay (one-item measure) | |||

Geographical Location | Geographical Location | |||

London | Rest of South East Britain | London | Rest of South East Britain | |

N | 158 | 185 | 158 | 185 |

Mean | 5.44 | 4.87 | 5.08 | 4.46 |

Std. Deviation | 1.181 | 1.377 | 1.335 | 1.437 |

Std. Error Mean | .094 | .101 | .106 | .106 |

Table 4: Independent Samples Test |
||||

Equal variances assumed | ||||

satisfaction with work content (one-item measure) | satisfaction with pay (one-item measure) | |||

Levene’s Test for Equality of Variances | F | 4.440 | 7.365 | |

Sig. | .036 | .007 | ||

t-test for Equality of Means | t | 4.097 | 4.133 | |

df | 341 | 341 | ||

Sig. (2-tailed) | .000 | .000 | ||

Mean Difference | .573 | .623 | ||

Std. Error Difference | .140 | .151 | ||

95% Confidence Interval of the Difference | Lower | .298 | .326 | |

Upper | .848 | .919 |

The average (SD) level of work satisfaction with work content of people of London is 5.44 (1.181) and of rest of South-East Britain is 4.87 (1.377).

The test statistics for the test is t(341) = 4.097. At 0.05 level of significance the p-value < 0.001. Since p-value < 0.05, level of significance, hence we reject the Null Hypothesis. Thus we can say that there statistically significant differences in the mean level of work satisfaction with work content between the residents of London and rest of South-East Britain.

Hence it can be inferred that the average level of work satisfaction with work content of the residents of London (5.44±1.181) is much higher than the residents of rest of South-East Britain (4.87±1.377).

The objective of the second analysis to test whether there is difference in the mean satisfaction with pay between residents of London and rest of South East Britain.

The hypothesis for the test:

Null Hypothesis: There is no difference between mean satisfaction with pay between residents of London and rest of South East Britain.

Alternate Hypothesis: There is difference between mean satisfaction with pay between residents of London and rest of South East Britain.

The mean (SD) work satisfaction with pay of residents of London is 5.08 (1.335) and of rest of South-East Britain is 4.46 (1.437).

The test statistics for the test is t(341) = 4.133. At 0.05 level of significance the p-value < 0.001. Since p-value < 0.05, level of significance, hence we reject the Null Hypothesis. Thus we can say that there statistically significant differences in the mean satisfaction with pay between the residents of London and rest of South-East Britain.

Hence it can be inferred that the mean satisfaction with pay of the residents of London (5.08±1.335) is much higher than the residents of rest of South-East Britain (4.46±1.437).

## Part 2a

The objective of the study is to find the relation between numbers of children in a household with working hours per week.

To test the objective the Karl-Pearson correlation between the two variables is used.

Table 5: Correlations |
|||

working hours per week | number of children in household | ||

working hours per week | Pearson Correlation | 1 | -.198^{**} |

Sig. (2-tailed) | .000 | ||

N | 343 | 343 | |

number of children in household | Pearson Correlation | -.198^{**} |
1 |

Sig. (2-tailed) | .000 | ||

N | 343 | 343 | |

**. Correlation is significant at the 0.01 level (2-tailed). |

Table 5 presents the correlation between the variables. The Karl-Pearson correlation between the variables is given as *r* = -0.198. Thus it can be seen that the correlation is negative. Hence with increase in the number of children in a household there is decrease in working hours per week. The correlation between the variables is weak, negative and linear. Moreover the correlation is statistically significant, p-value < 0.001 at 0.01 level of significance.

## Part 2b

The objective of the study is to find the correlation based on gender.

Table 6: Correlations |
||||

gender | working hours per week | number of children in household | ||

male | working hours per week | Pearson Correlation | 1 | .048 |

Sig. (2-tailed) | .549 | |||

N | 160 | 160 | ||

number of children in household | Pearson Correlation | .048 | 1 | |

Sig. (2-tailed) | .549 | |||

N | 160 | 160 | ||

female | working hours per week | Pearson Correlation | 1 | -.385^{**} |

Sig. (2-tailed) | .000 | |||

N | 183 | 183 | ||

number of children in household | Pearson Correlation | -.385^{**} |
1 | |

Sig. (2-tailed) | .000 | |||

N | 183 | 183 | ||

**. Correlation is significant at the 0.01 level (2-tailed). |

Table 6 presents the correlation between the variables.

For male residents the correlation *r* = 0.048. Thus the correlation is very weak, positive and linear. Thus with increase in the number of children there is very small increase in the number working hours per week for male respondents. Moreover the correlation is not-statistically significant, p-value = 0.549 > 0.05, level of significance.

For female residents the correlation *r* = -0.385. Thus the correlation is moderate, negative and linear. Thus with increase in the number of children there is moderate decrease in the number working hours per week for female respondents. Moreover the correlation is statistically significant, p-value < 0.001, less than *a* = 0.01, level of significance.

Thus, it can be seen that while working hours per week decreases with households having children, the same increases very slightly for males.