Please enable JavaScript.

Coggle requires JavaScript to display documents.

VIVA - Coggle Diagram

- - - - 10.1 understand the value of reflective practice and the need to record the outcome of such reflection to support continuous improvement
      - 10.2 recognise the value of multi-disciplinary reviews, case conferences and other methods of review
      - 10.3 reflect critically on their practice and consider alternative ways of working
      - 10.4 understand models of supervision and their contribution to practice
- - - - Reaction: Training managers use short surveys to gather data about participants’ responses to their training.
      - Learning: Participants complete an MCQ survey or quiz both before and after the training for training managers to determine how much knowledge has been acquired.
      - Application and implementation: The Phillips model doesn’t only collect data to find if the training worked or not; it also evaluates the WHY behind the success/failure of the training. It adds qualitative feedback to the data process to help organizations improve their training programs.
      - Impact: The model lets you analyze the impact of training content and other factors that contribute to participants’ final performance.
      - * Return on investment: Uses cost-benefit analysis to map impact data to tangible monetary benefits and a set of intangible benefits. Training managers can use this data as hard evidence to prove the value of training to the authorities.
  - - - Identify success cases.
      - Identify failure cases: Identify individuals or groups who have not achieved the desired performance improvements despite participating in the training. These failure cases represent the potential barriers or limitations of the training program.
      - Conduct in-depth interviews with both success cases and failure cases.
      - Analyze the data collected from the interviews to identify common themes, patterns, and trends. To determine the critical success factors as well as the factors that hindered the training effectiveness.
      - Action planning: Develop action plans based on feedback and recommendations to enhance the training program. These plans may involve refining the content, delivery methods, and support systems or addressing specific challenges faced by participants.
  - - - The Anderson learning evaluation model is a three-stage evaluation cycle applied at an organizational level to measure training effectiveness. The model is more concerned with aligning the training goals with the organization’s strategic objectives. ----- 1. Determining current alignment against strategic priorities. 2. Using your range of methods to assess any evaluate the contribution of learning. 3. Establishing the most relevant to purchase for the organisation.
  - - - Results: Check if the business objectives (such as greater productivity and fewer errors) linked to corporate training are met.
      - Behavior: Evaluate if employees apply their learnings from training to their everyday work.
      - Learning: Gauge the understanding of a topic and degree of skill development by taking pre and post-test measures and hands-on assignments. This helps understand if the training objectives were met.
      - Reaction: Understand learner satisfaction through feedback forms.
- - - - Amidst the accelerated integration of AI into the workplace, ethical considerations emerge as a pivotal competency for managers. The ethical dimensions of AI demand nuanced understanding and proactive decision-making to ensure responsible deployment and mitigate potential harm.
      - The imperative for ethical consideration in AI management is underscored by a growing body of evidence revealing inherent biases in AI algorithms.
      - Research by Buolamwini and Gebru (2018) demonstrated that facial recognition systems exhibited gender and racial biases, with higher error rates for women and individuals with darker skin tones. Such biases not only compromise the fairness of AI applications but also have profound implications for diversity, equality, and inclusion within organisations.
      - Moreover, ethical lapses in AI can erode public trust and lead to legal ramifications. High-profile cases, such as the misuse of AI in hiring processes resulting in discriminatory practices, exemplify the far-reaching consequences of neglecting ethical considerations (Mullainathan and Spiess, 2017). Managers, therefore, must be equipped to scrutinise AI systems for biases, champion fairness, and navigate the ethical intricacies of AI decision-making to uphold organisational integrity.
      - Ethical competency in AI management extends beyond bias detection. It involves a proactive stance in addressing the broader societal implications of AI, including privacy concerns and the potential impact on employment. As AI algorithms increasingly influence decisions in areas such as finance, healthcare, and criminal justice, managers must grapple with the ethical dimensions of these applications and ensure that AI aligns with societal values.
      - Change management as a competency extends beyond mitigating resistance; it involves cultivating a culture of adaptability and continuous learning within the organisation.
      - Technical Literacy
      - AI EU REGULATIONS
        
        The EU AI regulation aims to ensure trust and safety in AI by defining levels of risk and banning unacceptable risk applications. This is of direct relevance to any test developer and test user who involves AI within their process or system.
        
        The risk categories cover:
        
        Unacceptable risk: AI systems considered a clear threat to the fundamental rights of people will be banned. This includes AI systems or applications that manipulate human behaviour to circumvent users’ free will.
        
        High-risk: AI systems that cover: critical infrastructures; medical devices; systems to determine key citizen access; law enforcement and justice processes; biometric categorisation and emotion recognition systems.
        
        Specific transparency risk: Users should be aware that they are interacting with a machine. Chatbots, deep fakes and other AI-generated content will have to be labelled as such – users need to be informed.
        
        Minimal risk: AI systems will fall into this category if they present only minimal or no risk for citizens’ rights or safety. Applications such as AI-enabled recommender systems or spam filters will be given a ‘free pass’.
        
        Systems to determine access to educational institutions or for recruiting people are prime examples of high-risk applications. So too are any system processes that link to determinations for health provision or access to essential services (like utilities, finance).
        
        Designers and operators of these high-risk systems are now required to comply with strict requirements. These include: risk-mitigation systems, high quality of data sets, logging of activity, detailed documentation, clear user information, human oversight, and a high level of robustness, accuracy and cybersecurity.
        
        It may strike ADM readers that most, if not all, required elements of the new regulation parallel the expected quality methodology behind designing and using a good psychological test, especially when presented in a digital format.
        
        It was this aspect that led the EFPA Board of Assessment (BoA) to respond to EU AI regulatory proposals in August 2021. Whilst welcoming more guardrails on AI use in assessment systems, the BoA urged the proposed areas of risk be proportionately categorised for existing psychological assessment across Work, Education and Health domains. For example, the use of a simple summative digital scoring ‘algorithm’ on a well-being screening tool could classify the test as AI driven, and thus high risk.