PART 1: REFLECTION ON PROFESSIONAL SKILLS
1.1 My definition of reflection
For me, reflection is looking back at a concrete experience, analyzing the experience and looking forward from the experience, with the aim to understand and improve myself professionally. This can only be done when you take the time to reflect on an experience and think, feel, talk, write, etc. For example, questions like ‘what did I want?’, ‘what did I feel?’, ‘what did I think?’ and ‘what did I do?’, but also what did the other want, feel, think and do. I this way you can analyze a negative experience to avoid it from happening again (by learning how to act differently when a same situation occurs again) and/or you can analyze a positive experience to learn which actions and reactions provided for a positive experience/outcome.
1.2 Ten principles of good interdisciplinary team work
In a team, and especially in an interdisciplinary team, you have to collaborate, adjust and take your team into account. You cannot only think about yourself. Nancarrow et al.  state ten principles of good interdisciplinary team work, which include:
1. “Positive leadership and management attributes”
This means having a clear leader of the team, with clear direction and management. A good leader is not a dictator, but a democratic leader with shared power. A leader provides supervision, support and personal development and acts and listens to the team.
2. “Communication strategies and structures”
This includes ensuring that there are appropriate systems to promote communication within the team and requires individuals with communication skills.
3. “Personal rewards, training and development”
This incorporates individual rewards, morale and motivation, training and career development opportunities and learning.
4. “Appropriate resources and procedures”
This insures that appropriate procedures and structures are in place to hold up the vision of the service (communication systems, appropriate referral criteria, etc.)
5. “Appropriate skill mix”
This is about the ability to make the most of other team members’ backgrounds and having a full complement of staff that covers sufficient/appropriate skills, competencies and a balance of personalities.
6. “Supportive team climate”
This relies on a culture of trust, value contributions and nurturing consensus, which create and inter professional atmosphere.
7. “Individual characteristics that support interdisciplinary team work”
These individual characteristics include knowledge, experience, initiative, knowing one’s strengths and weaknesses, listening skills, reflexive practice and the desire to work on the same goals.
8. “Clarity of vision”
This includes having a clear set of values that drive the direction of the service and the care provided. It is about portraying a uniform and consistent external image.
9. “Quality and outcomes of care”
This requires a patient-centered focus and capturing and recording evidence of the effectiveness of care and using that as part of a feedback cycle to improve care.
10. “Respecting and understanding roles”
This relies on the sharing of power, joint working and autonomy.
I still have to work on communication and on individual characteristics that support interdisciplinary team work like knowing my strengths and weaknesses, asking questions and speaking out. A concrete example from this course that shows that I have to develop these specific skills is a big argument that I had with one of my team mates around the fifth week of the project. This argument arose because both of use did not speak our minds. I thought I was doing well in the project, but she thought I wasn’t and didn’t speak out about it. Once she did, it lead to a big miscommunication/ argument, which I did not really know how to deal with and that is why I did not speak my mind either. This proves that I have to work on my individual skills like asking questions and speaking my mind.
1.3 Positive, negative and learning experiences
In this quartile, I had to work in an interdisciplinary team. I have both positive and negative experiences from collaborating with students from another discipline. One thing that should be mentioned, is that I already have collaborated with different disciplines for different projects, including Business & IT, Electrical Engineering, Industrial Design, Mechanical Engineering and Industrial Engineering & Management.
Firstly, positive experiences that I gained from collaborating with Psychology students are getting to know how Psychology students think and work, what methods they use to accomplish goals and to learn the differences between Psychology students and Creative Technology students in general.
Secondly, negative experiences I gained from collaborating with Psychology students are clashing when using different methods (a discipline based problem) and clashing personalities that cause miscommunications and arguments.
However, I have learned from these negative experiences. I learned that different people and different disciplines work and think differently, and that you have to adjust to each other. You have to listen to each other, ask questions to each other and try to understand each other. I learned to reflect on miscommunications and arguments and where these miscommunications and arguments originated, to subsequently solve them.
Lastly, future goals I have for my own behavior to improve future interdisciplinary team work are keeping calm during disagreements and arguments, asking questions to (future) team mates to learn and understand how they feel (to avoid miscommunications and arguments) and to tell my (future) team mates more what I think and how I feel.
PART 2: REFLECTION ON PERSONALIZATION & IMPLEMENTATION
This part will provide thoughts on the collection of log data and health monitoring data, privacy and security issues, informed consent, integrity of data and data analysis, and the diffusion of innovation theory.
2.1 Collection of log data
Collection of the following log data is important to improve and personalize the prototype.
Firstly, to improve the prototype it is important to monitor if the program itself and the interaction between the system and user is logical and straightforward. In other words, it is important to monitor how users navigate over the application/interface. The goal of capturing this data is to find complications in user interaction with the system and possibly adapt the system according to these problems. If the system is not logical/straightforward, the user is not going to use the system. This is either because they do not know how to use the system and/or it is too time consuming to find out how they do have to use the system.
Secondly, to improve and personalize the prototype it is important to monitor when (at which points) and why people quit using the system. This includes monitoring after what time users drop out but also why users drop out. The goal of capturing this data is to provide insight into possible interaction complications and to provide insight into duration of use. Monitoring possible interaction complications and adjusting these problems accordingly can lead to an improved prototype. Monitoring duration of use can provide personalized experiences and personalized dialogue support/feedback.
Thirdly, to personalize the system it is important to monitor which environments are mostly chosen and thereby monitor which environments are favorable to the user. The goal of monitoring favorable environments is to provide personalized environments and environment recommendations.
Lastly, to personalize the system it is important to monitor when the application is used. This enables the creation of user profiles and usage patterns. In addition, it can personalize and enhance moments for persuasive triggers.
2.2 Collection of health monitoring data
Collection of the following health monitoring data is important to personalize the prototype. The goal of collecting this data is to provide users with user profiles, usage patterns and personal health records. It provides insight to the user about the duration of which they have exercised, when they exercised, why they exercised, etc.
Firstly, it is important to monitor whether elderly actually exercised. This provides the users with knowledge about whether they reached their personal health goal for that day.
Secondly, it is important to monitor when they exercised. This data is used to personalize the system and dialogue support. This data is also used to create usage patterns and to complement user profiles.
Thirdly, it is important to monitor why they exercised. This data is incorporated into the system’s dialogue support and serves as a motivation to the user. This data is also incorporated into the user profiles.
In addition, it is also important to monitor for how long they exercised. This data is mainly used to complement user profiles.
Lastly, heart rate is important to monitor. Elderly people should not be anxious or nervous to use the system. More important is that they do not get exhausted. This needs to be monitored and prevented. Monitoring weight is a possibility. However, the system is not aimed at obese elderly that have to lose weight.
Next, a reflection on privacy and security issues, informed consent and integrity of data and data analysis is provided.
2.3 Privacy and security issues
As stated in , complex data sets are a risk to personal data privacy. Privacy and security issues occur when new data is related to stored, anonymized data. The data collection for ‘Dementia in Motion’ is about private information. To protect personal data in the most confidential way, a data privacy protection strategy is needed. This strategy is an essential step to prevent harm to users. Several security protocols with respect to the collection, sharing, data-access, storage and protection, and (future) analysis of health monitoring-data are provided.
With respect to the collection of data, the following protocols are chosen. Firstly, the collection of data is anonymous or will be anonymized. As we are dealing with a limited data set, it is sufficient to anonymize the obtained data. Secondly, good data cleaning upon intake or upload is important to remove confidential data from data sets. Thirdly, it is important to run scanners that detect data on their drives that contain identifiers such as social security numbers. A social security number is a commonly used identifier with a characteristic format. Therefore it is easy to detect. In addition, it can be helpful to periodically conduct a “confidential data audit” to identify and delete such data.
With respect to the sharing of data, the following protocol is chosen. It is chosen to never share data sets by email or by a device such as a USB drive.
With respect to the access of data, the following protocol is chosen. It is chosen that a device that is accessing the master database or original data has specific, time bound privileges for accessing these data.
With respect to the storage and protection of data, the following protocols are chosen. Firstly, local, firewalled servers that are physically protected are the least risky option for data leaking. So, the data must be stored on a local, firewalled, physically protected server. In addition, the local device must have additional security layers such as whole disk encryption. Whole disk encryption prevents the device from being used or remotely accessed by others. Moreover, theory-driven inquiry may improve privacy protection.
With respect to (future) analysis of health monitoring data, the following protocols are chosen. Firstly, the data that is analyzed only contains the few critical variables that are needed for the analysis. The data does not contain identifiers or personal information. Secondly, the data is analyzed on a secure device. The specific file for analyzation is encrypted. That data file is destroyed once analyses are run. Parameters can be documented for replication, but the actual data file is destroyed. A scan is run to find and remove any additional copies. If original data must be retained, it is stored on a server with physical and IT security.
2.4 Informed consent
According to , informed consent is essential for minimizing harm to research participants. People give consent to researchers to use their data for certain purposes. When obtaining informed consent is appropriate, the APA recommendations (8.02) specify that researchers inform participants about, for example, the purpose, duration and procedures of the research, their right to decline to participate and withdraw once they have chosen to participate, foreseeable consequences/potential risks, and contact information. In addition, researchers conducting intervention research that involves the use of experimental treatments should clarify, for example, the experimental nature of the treatment and available treatment alternatives when someone declines to participate or when one wants to withdraw once one has chosen to participate. This information needs to be thought about and given to elderly to obtain informed consent. This informed consent can be obtained by filling in a form providing the information stated above or, for example, through a video, which is more suitable for elderly. As the target group is elderly, the way of providing information must be seriously thought about, to prevent confusion and indistinctness.
2.5 Integrity of data and data analysis
In this section, critical issues regarding the appropriate use of big data in research and practice are provided.
2.5.1 Identification, Acknowledgement, and Minimization of Potential Harm/Risks of Big Data
Researchers have an ethical responsibility to identify, acknowledge and minimize any potential risks or harm that is caused by their study. Moreover, they have to make sure that these risks are known to the participants.
It is stated in  that a single data set on its own carries little risk, but that a single data set can pose meaningful risks when it is combined with other data sets. However, this is avoided by several security protocols that are explained in part 2.3. In addition, it is stated in  that applications of algorithms and machine learning form potential risks, as algorithms can hard-code biases into systems. This can cause a risk for, for example, automated selection systems. However, this is not the case for ‘Dementia in Motion’.
2.5.2 Identification and Disclosure of Sampling Issues
A representative sample is not necessarily a large data set, as large data sets are exposed to problems associated with nonrandom samples and biases in explanation. It is necessary to make clear which sampling methodology is used and sampling-related limitations need to be acknowledged.
2.5.3 Maintaining Standards of Measurement Quality
As stated in , an important aspect of big data is the quality of the measures comprising big data.
Data from high tech devices needs to be demonstrated as reliable and valid. An example of these high tech devices are wearable sensors. The reliability and sensitivity of these sensors need to be known to researchers to prevent them from making false inferences. In addition, evidence supporting the quality of measures should be routinely provided. However, ‘Dementia in Motion’ does not use wearable sensors. Moreover, ‘Dementia in Motion’ only uses the heart rate monitor that is built into the home trainer, which is fairly reliable.
2.5.4 Ensuring Data Quality/Veracity
Efforts have to be made to ensure data quality/veracity. Questionable data points/outliers are identified and corrected or possibly discarded. Limitations to veracity need to be acknowledged.
2.5.5 Making Appropriate Claims Based on Data Analysis
When it comes to interpreting results, researchers need to be cautious. Potentially spurious correlations need to be acknowledged. Replication and cross-validation need to be performed before drawing conclusions. When ‘Dementia in Motion’ works for some, it might not work for others.
Diffusion of innovation theory
This section provides a short reflection on how the implementation of the prototype is organized by applying the diffusion of innovation Theory , . The dynamics of innovation diffusion are applied on the prototype. Motivations are given about how the adoption of the prototype is optimized.
Dynamics of innovation diffusion applied to the designed technology:
1. “Perceived benefit”
The innovation combines virtual reality with exercising for elderly with mild dementia.
This is not done yet by current technology. According to a user test with a target group representative, the added element of virtual reality makes the technology fun to use.
The innovation fits into the current mindset of the user. It is easy to think that this is not the case for elderly users, but based on interviews with target group representatives the previous statement is supported.
The innovation causes little frustration and can be used in the environment of the user. However, the system requires space, as a home trainer is not very small. Though, local adaptation is possible. For example, a user can sit on the couch instead of on the bicycle when using the system. This is also to prevent exclusion of elderly that cannot ride a bicycle anymore due to leg problems. In addition, the interface and system overall is created to be as easy and clear as possible. This is to avoid confusion and frustration.
The user is able try the innovation before adopting it. The user is able to try the innovation in a local technology shop before buying it. In addition, the system can be returned within the trial period of 30 days when it does not please the user. This is all aimed at preventing ‘mandatory’ adoption of a new, unfamiliar technology.
The user is able to first observe someone else while using the technology before making the decision to use it themselves. This is correlated to the trialability aspect in a way that they can order the technology, let someone else demonstrate it first, and if the elderly user does not like the technology, return it. In addition, possible advertisements can help to increase a sense of observability.
...(download the rest of the essay above)