Presentation agenda - ideas sharing with dissertation supervisors

Presentation agenda - ideas sharing with dissertation supervisors:

1. Awareness of the main project barriers with regard to supervisees:
1.1. Time constraint, time poverty and time management
1.2. Learning mindset
1.3. Intellectual capability
1.4. Academic resource availability
1.5. Misunderstanding on various dissertation project practices

2. Diversity of supervisees' profiles and the supervision issues involved

3. Advice vs instruction to supervisees; support to supervisees  vs complaints from supervisees

4. Supervisors' feedback practices at various stages of the dissertation project: oral and written ones

5. Assessment and benchmarking practices

Trouble-shooting for Excel regression reporting: some advice

Trouble-shooting for Excel regression reporting: some advice.

Some Independent Study students report on the following issues as related to Excel regression reports when conducting multiple regression analysis:

1. t-value and p-value figures are contradictory: I suggest that you examine the figures of p-values. If the figure is 2.3E-3; that means the figure is 0.00.23. You could choose the comma format, allowing for more decimal points to let Excel shows the actual figures in a way that you could understand better.  My experiment 1 result below is indicative (y variable figures are all the same):

Experiment 1: variable y value is constant

Y-variable	X1	X2	X3	X4	X5
1.2	21	14	0	23	47
1.2	23	15	1	24	48
1.2	18	16	0	25	48
1.2	19	17	1	26	50
1.2	22	18	0	27	51
1.2	17	19	0	28	52
1.2	21	20	0	29	53
1.2	20	21	1	30	50
1.2	20	22	1	24	55
1.2	22	23	0	32	56
1.2	25	24	1	20	55
1.2	25	25	1	34	58
1.2	20	12	0	23	47
1.2	21	12	1	19	60
1.2	20	15	1	26	45
1.2	18	17	1	21	42

	Coefficients	Standard Error	t Stat	P-value
Intercept	1.2	1.88E-31	6.4E+30	2.1E-304
X1	-1E-33	7.76E-33	-0.13156	0.897945
X2	-1.1E-33	5.22E-33	-0.21339	0.835314
X3	2.48E-32	3.17E-32	0.781673	0.452516
X4	1.32E-33	4.6E-33	0.287302	0.779746
X5	1.81E-34	3.67E-33	0.049329	0.961629

Revised report format

	Coefficients	Standard Error	t Stat	P-value
Intercept	1.2000000	0.0000000	6,396,897,226,530,190,000,000,000,000,000.0000000	0.0000000
X1	-          0.0000000	0.0000000	-                                                                                                                         0.1315552	0.8979452
X2	-          0.0000000	0.0000000	-                                                                                                                         0.2133856	0.8353138
X3	0.0000000	0.0000000	0.7816735	0.4525163
X4	0.0000000	0.0000000	0.2873024	0.7797462
X5	0.0000000	0.0000000	0.0493285	0.9616288

If the p-value is #NUM! while the t-value is very large, that may means that the data set has some extreme figures, e.g. all x1 variable figures are the same, see my experiment 2 result below). [Note: the explanation of #NUM! from Microsoft can be found in this link].

2. If the t-value is very large, then what can be done? Try to check the underlying data file figures to see if there are outlier figures, which explain why this is the case; or, it is due to data entry errors. Another way to deal with that is to use a record as based record to calculate sensitivity figures. Sensitivity figures (in % terms) tend not to vary so significantly as the absolute figures. Anyway, even if the t-value is very large, e.g. 1,000, it is still possible to conduct a hypothesis testing by comparing the sample t-value of 1,000 with the critical value of 1.96. At the same time, students could also consider to revise or introduce some new quantitative tests to explore the data set further.
3. It is also useful to revise the formula design to explore the underlying data set further.
4. If all efforts to cope with the technical problem fails, try to stick to one set of figures for analysis, i.e., the t-value of the p-value. This is the means of last resort. This step has risk, however; see point 5b below.
5a. If the b value is 0 and the standard error figure is also 0 for an x-variable, my experiment finding shows that the variable figures for this x-variable in the file is a constant value.  See my experiment result as follows:

Experiment 2 : variable x1 is a constant value.

Y-variable	X1	X2	X3	X4	X5
21	1.2	14	0	23	47
23	1.2	15	1	24	48
18	1.2	16	0	25	48
19	1.2	17	1	26	50
22	1.2	18	0	27	51
17	1.2	19	0	28	52
21	1.2	20	0	29	53
20	1.2	21	1	30	50
20	1.2	22	1	24	55
22	1.2	23	0	32	56
25	1.2	24	1	20	55
25	1.2	25	1	34	58
20	1.2	12	0	23	47
21	1.2	12	1	19	60
20	1.2	15	1	26	45
18	1.2	17	1	21	42

	Coefficients	Standard Error	t Stat	P-value
Intercept	9.617407577	6.683169194	1.439048945	0.177978
X1	0	0	65535	#NUM!
X2	0.145132095	0.197826159	0.733634501	0.478517
X3	0.68778005	1.214593666	0.566263491	0.582583
X4	-0.032302779	0.178243671	-0.181228194	0.859485
X5	0.17517698	0.132191307	1.325177764	0.211979

Anyway, it is useful to examine the underlying data file to figure out what happens in the data pattern. It also points to the need to make further refinement of the research design, notably on the regression formula setting. One needs to especially review the choice of x1 in the formula as shown in my experiment above.
5b. When the data set consists of records with such extreme values, the Excel calculation may run into some difficulties, resulting in certain calculation errors. On this point, I am not able to verify however. I do feel that more fundamentally, the formula design and data gathering approach need to be revised or refined in this case.


Two additional final viewpoints:
Point 1: It is very rare that the problem of Regression reporting is due to Excel bugs. And, after you resolve the problem, try to justify your revised research design in your Independent Study report.
Point 2: Although I do not have the technical knowledge to answer all Excel problem, I feel that we do not need multiple regression analysis to study a data set with a large amount of extreme figures. Extreme data patterns, in this case, can simply be studied with more basic descriptive statistics, e.g., scatter diagrams, in my view.

A multidimensional data analysis of questionnaire survey data on the sharing economy perceptions in Hong Kong using the Excel pivot table function

A multidimensional data analysis of questionnaire survey data on the sharing economy perceptions in Hong Kong using the Excel pivot table function

JOSEPH KIM-KEUNG HO

 Independent Trainer

Hong Kong, China

Dated: May 16, 2018

Abstract The Excel pivot table function (EPT) used for multidimensional data analysis (MDA) should be considered as a research decision support system (DSS) so that its value as a research tool can be better realized. This paper provides some theoretical clarification on this proposition from the writer. Specifically, it  provides an illustration on how the multidimensional data analysis is done with  a EPT application on a data file on a 2015 questionnaire survey conducted by the writer with regard to the topic of perceptions on the sharing economy in Hong Kong.

Key words: Excel pivot table (EPT), The sharing economy perceptions survey data, Multidimensional Data Analysis (MDA), Research Decision Support Systems (DSS).

Introduction

The Excel pivot table function is a handy data analysis tool for multidimensional data analysis. Ho (2018) further postulates that it could be conceived as a research decision support system. In this paper, the writer employs the Excel pivot table function to study a survey data set on the Hong Kong sharing economy perceptions gathered by the writer in 2015 (Ho, 2015). The objectives are to: (i) demonstrate the research decision support system practice with the Excel pivot table and (ii) produce additional findings on Hong Kong sharing economy perceptions based on the survey data set of Ho (2015). As such, the paper should be of interested to those who study the subjects of research methods and the sharing economy.

Using Excel pivot table function for multidimensional data analysis

The Excel pivot table function utilizes a flat data file in Excel, with the first row consisting of field names and the subsequent rows as records. From the point of view of the pivot table function, each field name indicates 1 dimension of the topic to be examined. Thus, with a number of fields, the Excel pivot table function promotes a multidimensional view of the topic under review. In essence, the Excel pivot table function enables a convenient way to conduct multidimensional data analysis on the researched topic. More than that, due to its interactive interface that allows its user to produce multiple 2-dimensional-table views on the underlying data file, this pivot table function enables its user to explore and exploit the data file to generate and verify information in the analysis process. Therefore, the Excel pivot table function has been recommended by Ho (2018) to be conceived as a research decision support system, and not merely as a tool for the construction of tables as descriptive statistics, which are often included in the chapter on "findings and analysis" in a typical academic dissertation report. The Excel pivot table is a decision support system as it consists of (i) a database (i.e., the flat data file), (ii) a model base (i.e., a pivot table template with a set of dimensions on the researched topic) and (iii) an interactive and usable interface (i.e., the Excel pivot table construction interface which allows for the creation of multiple 2-dimension table views on the underlying data file). Because of that, Ho (2018) encourages researchers to adopt decision support systems methodologies and practices to work with the Excel pivot table function. The following example on the application of the Excel pivot table function to a survey study of the Hong Kong sharing economy further illustrates this research decision support system notion. This account comprises two parts: the first part is a brief and updated introduction on the sharing economy topic and the second part describes the actual Excel pivot table function application on the survey data on the sharing economy perceptions as reported in Ho (2015).

The main ideas underlying the topic of the sharing economy in brief

Albeit the "widespread ambiguity" of the term "sharing economy" (Frenken and Schor, 2017), it is informative to conceive it in certain ways, as, e.g., "consumers granting each other temporary access to under-utilized physical assets ("idle capacity"), possibly for money" (Frenken and Schor, 2017) and "a new culture of sharing.... in which people make their belongings accessible through online networks" (Bucher, Fieseler and Lutz, 2016; Zhang, Gu and Jahromi, 2018). The prime characteristics of the contemporary sharing economy are (i) "stranger sharing" and (ii) the sharing of "shareable goods"[1] (Frenken and Schor, 2017) while the concept of "sharing sentient beings" being more controversial (Yau, 2018). Böckman (2013)  further examined the "sharing economy" phenomenon in terms of the societal, economic and technological drivers of it. Beyond that, Ho (2015) identifies six related study areas on the "sharing economy" topic, namely, (i) the nature of sharing, (ii) subcategories of the sharing economy, (iii) Information technology (IT) platforms and infrastructures, (iv) business models and strategies, (v) impacts and stakeholders' concerns, and (vi) recommended government policies and regulations. This updated account of the sharing economy topic, which is necessary brief, primarily serves to facilitate comprehension of the multidimensional data analysis findings to be presented in the latter part of this paper.

An application of Excel pivot table function to study survey data on sharing economy perceptions in Hong Kong

The survey data to be studied with the Excel pivot table function comes from the 2015 Facebook-based questionnaire survey conducted by Ho (2015) on perceptions of the sharing economy in Hong Kong. By using the Excel pivot table function to study this set of survey data, the writer gives a demonstration of how the Excel pivot table function can be employed as a research decision support system (Ho, 2018) as well as offers additional analysis findings on the 2015 survey of the writer's sharing economy perceptions study. The 2015 questionnaire survey, conducted in August 27 to 30, consists of 18 questions , see appendix 1. The initial questions learn about the profiles of the survey respondents while the subsequent questions record the respondents' perceptions on topics related to the sharing economy practices and prospects in Hong Kong in 2015. By cleansing the survey data, the writer is able to conduct a multidimensional data analysis with the Excel as the research decision support system on them.

.... for further information, please download the pdf version.

---

[1] Shareable goods are "goods that by nature provide owners with excess capacity, providing the consumer with an opportunity to lend out or rent out their goods to other consumers" (Frenken and Schor, 2017).

A pair of management concerns and theoretical framework: an example

A pair of management concerns and theoretical framework: an example


Diagram 1: On management concerns

Diagram 2: on theoretical framework

Making overall assessment comments in the "Findings and Analysis" chapter in applied business research report

Making overall assessment comments in the "Findings and Analysis" chapter in applied business research report: 2 groups of suggestions

Most students have no difficulties to make findings statements based on individual research method used; however, to come up with some assessment comments based on a synthesis of all the research methods findings and based on the theoretical framework (as produced via the agile literature review approach) is easier said than done. Here are my suggestions on how to do so:

Group 1: factors and linkages analysis
1. Identify (i) the most critical factor(s) and linkage(s) from all the research findings and (ii) the relatedness of these critical items (i.e., critical factors and linkages) as a kind of "critical path".

2. Identify a set of critical factors with linkages and comment on the combined effect of it (the set of critical factors) on other domains of your theoretical framework (e.g., the outcome domain of your theoretical framework).

Group 2: emergent properties analysis
By considering all the research findings together, you form an overall impression on the problem-situation and make some evaluative comments on emergent properties of the system(s) under review, notably the company studied. These emergent properties can be associated to the following management topics:

1. Strategic drift 1/ strategic drift 2
2. Strategic misalignment (especially with I. Ansoff's rating forms)
3. Strategic leadership
4. Obstacles to change and learning
5. Organizational types of H. Mintzberg
6. Types of management theories (R.E, Miles)
7. Existence of certain wicked problems, that cannot be addressed / solved with piecemeal approaches

Housing imaginations assignment 2 - some guidelines

Housing imaginations assignment 2 - some guidelines

1. Critically discuss the assertion by Doreen Massey (1992) that homes, like all ‘places’, are open to, and created by, the social relations which extend beyond them.

2. To what extent does the concept of ‘place’ provide a useful; basis for understanding the dynamics and conflicts surrounding home-places?

3. In what ways is the housing imagination gendered an d classed?

Essay structure:

Part 1: explain (i) the main viewpoint of the statement (e.g. the intended analytical usage of it), (ii) the underlying Housing Imaginations concepts involved and (iii) how these concepts are related to each other as suggested by the statement.

Part 2: Discuss the statement in the following sequence:

2.1. An elaboration with the supportive standpoint (e.g. sociological perspective (s) on the statement would be as follows, with the following examples for illustration

2.2. An elaboration with the opposing standpoint on the statement would be as follows, with the following examples for illustration.

2.3. A discussion on possible blind spots on the standpoint as endorsed by the assignment statement.

Part 3: Concluding remarks

3.1. Explore any room for coping with the tension involved in the debate (i.e., between 2.1, 2.2 and 2.3).

3.2. Reflect on the discussion and offer your personal viewpoint on the topic.

3.3. Based on 3.3., suggest some recommendations that can be made.

*** pls ensure that you make use of some relevant HI concepts in your discussion.

Chapter structure on Conclusions and recommendations with the agile literature review approach (ALRA):

Chapter structure on Conclusions and recommendations with the agile literature review approach (ALRA):

Dissertation project title
Executive summary
Acknowledgement
Table of contents
List of figures and tables

Chapter 1: Introduction
Chapter 2: Literature review

Chapter 3: Research methods
Chapter 4: Findings and analysis

Chapter 5: Conclusions and recommendations

5.1. Chapter introduction

5.2. An evaluation of the actionable value of the main research findings: some concluding comments

5.3. Recommendations
5.3.1. Recommendations at the strategic level
5.3.2. Recommendations at the tactical level
5.3.3. Recommendations at the operational level

5.4. An evaluation of the academic value of the dissertation project work

5.5. Concluding remarks on the extent of achieving the research objectives

5.6. Some reflexivity-related comments on the dissertation project practice performed.


Chapter 6: Personal development

References

Appendix