Theory
Project Title: Is transformational management made easier by understanding an audiences’ personality type(s)?
Ontology: Relativism
Epistemology: Social Constructionism
Methodology: Quantitative Mixed Mode
Thesis statement
Different personality types have a different outlook on digital transformation and we believe understanding these differences will aid the implementation of transformational change management programmes.
Abstract
This proposal involves discovering how Jung, 1990 personality types are an indicator as to how they accommodate change, specifically technical change within transformational change programmes. The goal is to show that certain personality types are more accepting towards technical change and as such more receptive within transformational change programmes Rogers, 2016.
This will be done through a questionnaire distributed via a variety of channels, anonymously as to encourage more honesty upon completion. The data will then be reviewed to identify any patterns which emerge focusing on the Millikin, 2017 Kano crosstabs to identify versatility of peoples’ attitude towards change.
Introduction
Over the last 10 years’ transformational change has become more common with the introduction of digital transformation change programmes now effecting most industries, with this increase the amount of potential for organisational culture impact has increased.
When considering digital transformation Rogers, 2016 outlines that we need to consider strategy, culture, customer, operations and technology. Whilst each element is important the people within “culture” are considered the most complex, Westerman et al., 2014 identified that to migrate any workforce from where they are now to where they will be being a substantial change and requires the appreciation for the different types of people involved.
Looking at alternative areas of business management namely marketing we commonly segment our target audiences into different groups as to support the targeting of those groups in an individual way whilst maintaining a cost effective operation. This method of targeting aids in the development of the audience and the targeting of a message to that audience to increase their adoption or conversion as required.
Whilst useful for communication purposes what support would it give in regards to an internal audience who are undergoing a transformational change? Is it detailed enough or do we need to consider a deeper level of segmentation and review personality profiles which are becoming more commonly used within industry today, those such as Myers and Myers, 2002 are becoming standard practice for recruitment so could the insights gained from understanding a person’s personality profile have an affect upon a person’s acceptance of change?
This leads onto the question of if transformational management is made easier by understanding the audiences’ personality types in further detail than just the appreciation that there is a difference? To understand this question it is our intent to conduct a review of different personality types and their perception towards technical change.
This literature review follows this train of thought and investigates the published works surrounding segmentation, digital change and personality profiles. It is broken down into idea themes, each theme is then broken down into sub themes which include evidence on the theme, identified arguments and examples.
Methodology
In order to understand a population’s behaviour (Aberson, 2010) towards changes in technology we intend to conduct quantifiable research through a single web based closed questionnaire (van Gelder et al., 2010). This is felt to give the most accurate data to build this study upon as it has the potential for the widest distribution which should identify if there are differences or trends between any of the personality types and their acceptance of technology.
Whilst this would be primarily web based there would also be paper copies distributed to a selection of different audiences as to ensure a good coverage of those with easy access to technologies and those without as described in section 2. This turns the model into a mixed mode questionnaire as described by Hohwü et al., 2013 but whilst it will use different methods for data collection all data will be entered into the same central data set.
The questionnaire approach has been selected as there are different elements (personality, demographic, technical acceptance) or as Tyler, 2007 describes those categories that need to be identified (described in section 5.3 below) as to see if there is a correlation with the different categories. There is usually a potential bias with questionnaires as people commonly lie due to social desirability (McLeod, 2017) however this is intended to be mitigated by making the entire study anonymous (Schleyer and Forrest, 2000) so there is no social pressure on the participant to answer a certain way. However, this in itself brings about the issue that someone could repeatedly answer the questionnaire which would taint the data, this is addressed by Chesney & Penny, 2013 and whilst is possible the probability of this being done to scale is very low as such the probability of enough duplicate answers to damage the data is minimal.
Alternative methods are available but due to the identification of personality types and their acceptance of technologies there is a high probability that any identifiable method would result in inaccurate data due to the individual’s desire for social positioning as explained by McLeod, 2017. Immediate anonymization was considered whereby the personal data would be collected but immediately hashed so it was unidentifiable separating the contact name and email from the results supplied and whilst this would lead to a separation of the data from the identifiable set the perception of those answering the questionnaire would still be tainted by the supply of personal data. Also the separation of these data sets would mean the benefits which could be obtained by holding the identifiable data would be lost, such as being able to contact any outlier data sets to investigate the reason why etc.
Questionnaire
This will be broken down into three sections as shown in figure 3 below.
Figure 3: Study breakdown by elements for personality vs technical change vs demographic
The Personality type questionnaire is based upon the OEJT (Open Extended Jungian Type Scales 1.2) (Jorgenson, 2017) this will identify the individual into one of the 16 personality types (Cattell, 2017) as shown in figure 3. This questionnaire is made up of 32 pairs which are connected with a 5 point scale, as shown in figure 4 within each pair you mark (as shown with the X below) where in the scale you feel you fit.
Figure 4: example pairs question with 5 point scale
Upon completion of this questionnaire this can either be manually processed using the formulas supplied by Jorgenson, 2017 or automatically processed through the website and SQL logic or an Excel document. It is our intention to key all responses through the website as to build a single consolidated source of this data.
The second stage is the demographic information (Salkind, 2010) which is captured in ranges of 5 years from below 18 to above 70 which gives 12 options on a single question. This range is set on the assumption that there will not be a major personality type change or a technical acceptance difference within a 5-year difference however there could be within a 10-year separation.
Figure 5: age range with 5 year separation
The third is a question of perception (Chau, 1996) so asks questions based upon comfort of using technology within certain situations this is measured using kano analysis (Millikin, 2017).
Figure 6: perception based question example.
This could also be measured using statements with a measurement of strongly agree to strongly disagree however according to Holroyd, 2012 there is an increased chance of implicit bias occurring in the formation of the statements. Whilst the detail of this questionnaire are still being finalised the outline is shown in appendix (A) as well a low fidelity mock-up of the website is included in appendix (B) showing how it could be structured once the questions have been fully formed.
Strengths and weakness for this approach
Questionnaires have been used throughout research history to varying degrees of success according to Goodman, 1997 the common reason for their efficiency is based upon the quality of questions being asked and the size of the sample which it is asked of. That being said Song et al., 2015 highlights that questionnaires can lead to inaccuracies if not designed well, so much so that they conclude that “Questionnaire design is more of an art than a science” Song et al., 2015 due to this a number of localised tests with small receptive audiences will be undertaken and feedback sought prior to the full questionnaire being released to a wider population.
Figure 7: SWOT analysis for approach
The biggest identifiable weakness to this approach is the sample size if it is not adopted and distributed by my social networks the response rate will be weak and whilst this could be increased through expanding the time for collection, this is not a feasible option within this study timescale.
The length of the questionnaire also leads to the possibility of a drop out whereby someone exits the questionnaire prior to completion this would lead to the data not being collected. The site will have some basic analytics enabled as to know how many people opened the page and how many people completed the questionnaire this will allow us to report on the drop off rate if needed.
Rationale
This approach was selected due to a number of factors above those listed in the Methodology section.
Firstly, to understand how to influence the transformational management (Song et al., 2015) of an enterprise organisation the sample size needs to be large enough to be representative of a similar scale to that of a large organisation. Whilst it does not need to be the same size it needs to be of an appropriate sample to bring about enough confidence in the data as to utilise it during a period of organisational change.
Secondly to understand a personality group effectively a reasonable level of responses for each personality type is required as to identify abnormalities (Monkey, 2017) as such the questionnaire approach gives the opportunity of volume. The exact sample required for this cannot easily be predicted as it depends on the sway of the answers, for example if all 10 respondents for personality type INTJ (Analytics, 2017) all show strong acceptance of technical change this may be enough to stop collection however if 6 show a strong acceptance but 4 show a minor acceptance then the sample needs to be increased until the data clearly shows in the positive or the negative. If all are in the positive the larger the sample size collected then the better the confidence level in the results will be.
Finally, the demographic sample needs to be sufficient for each personality type as to avoid any age related bias, again this requires a good volume of respondents as to identify any outliers from the standard responses.
Ethical considerations
Given the decision to conduct the questionnaire on an anonymous (Birnbaum, 2004) basis by not collecting any identifiable information the ethical implications are at a minimum.
Given this is a questionnaire approach the easiest method would be to use a free survey tool such as survey monkey, zoomerang or surveygizmo however most survey tools collect some data from free surveys for their own purposes and whilst there is no personal data (Commissioners Office, 2017) being collected this means there would be no detrimental impact upon the end user.
That being said the preference is to use a service whereby the data is controlled from beginning to end such as our own server or web hosting where the database can be secured as to avoid any misappropriation of the data. Any logs on this server which hold the connection info such as IP address would be cleared every 24 hours and whilst an IP address is not identifiable to a person it maintains the highest level of animosity if these are cleared.
Data analysis approach
Given the data is split into three sections for collection the first phase of the analysis must follow the same separation
Personality types
For personality types the Open Extended Jungian Type Scales (Jorgenson, 2017) will be used this provides the logic for interpreting the questions into each of the 16 personality types, this has been tested in excel and it gives identical results to the more well-known Myers et al., 1996 test which is reassuring as both are forked from the original Jung, 1990 personality type research.
For mainstream distribution, it is intended to code this directly into a website so the data for personality type will be processed instantly adding the identified personality type directly alongside a simple graphical representation of the results (Gardner, 1996). This covers my public contribution element of the data as to encourage increased participation and as such increase the sample size (Monkey, 2017).
Perception
Originally this was only going to focus on the Siegel, 2008 TAM approach which asks about the perception of a product before its use which by using the separation of activity at home and activity at work would identify those who use a technology because they want to vs those who use it because they need to.
However, from further research the Kano model (Zacarias, 2017) offers clearer understanding into the satisfaction by separating this into three key areas attractiveness, performance and must-be. By using this separation, we hope to identify not only if there is a correlation between those personality types who will accept technical change but also show a model for engagement with the different types as described by Zacarias, 2017 based upon what they expect (must-be), want (performance) and desire (attractive) from which as Pilkington, 2013 explains communication and engagement plans could be drawn up to actively encourage these audiences to participate in organisational change.
The below examples of the kano model (figure 8 and 9) show customer need, the two different models where the must-be (threshold attributes) and the desire (excitement attributes) have moved position based upon the audience expectation.
Figure 8: Kano model example 1
Figure 9: Kano model example 2
From these examples, we can see that figure 8 audience requires a higher level before they are excited by a product but also have lower expectations of what must be included whereas figure 9 highlights that the base product must include more functionality however they are excited by a lower level of improvement comparatively.
When these same examples are applied to our research question it may be the case that one of the personality types has a very low technology use within the must section and is not excited by technology and as such would be harder to encourage to adopt a technical change when compared to someone who is excited by technology changes.
After that we will also be analysing the data using Excel and SPSS (Prvan et al., 2002) to see if there are correlations between certain personality types and the technical acceptance. The approach to be followed is known as the Kano model (Tontini, 2007) which looks at the data once collected and segments it into different categories based upon its positive or negative virtues from this the lines of attributes move to form the structure as shown in figure 8 and 9 where the lines of attribution are in different positions based upon the collected data. This will be processed through a series of crosstab reports as to identify the % of people from different personality profiles who fall into the different sections.
Demographic
Given the demographic collection (Bernhardt, 2013) is quite limited as to maintain animosity of the participants and will only be asking for an age range and gender at birth. It is appreciated that there may be more ways to interpret this data once it is collected however that is not the purpose of this study so the demographic element does not need to be extensive only to support or dispute the hypothesis of age being a factor for personality types accepting technology change.
Data presentation plan
Evergreen, 2013 suggests it is best to present data in as pure a form as possible as well as interpreted, as such the data will be presented in a few different stages. Firstly, in its purest form of the data whereby the results will be displayed question by question with the responses to each for example the Demographic question set shown below will be shown in the first instance just with numbers of participation for each field as shown with the fictitious numbers in figure 10 below.
Figure 10: table example of demographic data