Learning objectives serve as guiding beacons, illuminating learners’ knowledge acquisition path. They determine what students should know, understand, and do to complete a course. These objectives are not just pedagogical compasses; they also form the foundation upon which course content, assessments, and instructional strategies are built.
I learned about the magical power of learning objectives when I created my first course. They are the cornerstones of a successful educational experience, aligning educators’ aspirations with students’ achievements.
The motivation behind this experiment lies in my quest to discover the Best AI for Learning Objectives. In this age of AI, where machine learning models can generate text with remarkable fluency, can they be trusted to craft meaningful, measurable, and actionable learning objectives?
To answer this question, I am on a journey to test the mettle of three AI tools: ChatGPT, Bard, and Bing. My goal is to evaluate their proficiency in generating learning objectives for an online course, specifically in this post, one focused on the intricacies of on-page SEO. The results of this experiment not only shed light on the AI’s aptitude for instructional design but also offer insights into the ever-advancing realm of AI-assisted education.
Learn more about Exploring the Future: Instructional Design and AI.
Join me on this exploration of the capabilities of AI in the realm of learning objectives. Together, we will uncover whether AI can assist instructional designers and educators meaningfully.
Table of Contents
This post is fairly long and detailed. Therefore, I’ll provide TL;DR and a more detailed summary initially.
TL;DR – Best AI Learning Objectives?
In this experiment, I compared AI tools (ChatGPT, Bard, Bing) to generate learning objectives for an SEO course. Bard initially struggled, while ChatGPT improved. Bing excelled with clear, measurable objectives. For practical use, combining insights from all three AI tools seems ideal. The experiment revealed surprising variations in results!
Summary: Which AI is Best for Generating Learning Objectives?
Which is the best AI Learning Objectives? This experiment is part of our series to explore AI’s capabilities in the instructional design field. This time, we try learning objectives: ChatGPT vs. Bard vs. Bing – all competing to be the best creator of learning objectives.
The experiment focused on creating learning objectives for an online course centred on on-page SEO, a crucial topic in the digital world. The outcomes were assessed using a comprehensive matrix based on key criteria: alignment, specificity, measurability, achievability, relevance, and outcome-based orientation.
The Results
- Bard: While Bard initially struggled to produce clear learning objectives, it eventually improved with hints from the prompts. Its final attempt provided measurable objectives.
- ChatGPT: ChatGPT had its own set of challenges but improved with adjusted prompts. Its second attempt yielded better results, although some objectives needed more specificity.
- Bing: Bing emerged as the clear winner from the start, consistently delivering well-structured, clear, and measurable learning objectives that adhered to standard formatting and best practices.
Overall Assessment
- Bard’s initial attempts affected its overall performance but improved with additional prompting.
- ChatGPT’s second attempt was an improvement over the first but still had room for enhancement.
- Bing demonstrated excellence throughout, providing the most suitable learning objectives for the SEO course.
Conclusion – Best AI for Learning Objectives
In the end, the experiment highlighted that with slight adjustments to the prompts, all three AI tools could generate suitable learning objectives. However, Bing excelled consistently, producing clear and measurable objectives. The varying outcomes also emphasized the importance of having a diverse set of AI-generated ideas to manually craft the best learning objectives.
The lesson learned from this experiment is the potential of combining insights from different AI tools to create a medley of learning objectives that align with one’s personal style and course requirements. The quest for the best AI for learning objectives ended with the recognition that a blend of AI capabilities can yield the most effective results.
The question remains: What is your best AI for Learning Objectives? Your feedback and ideas for future AI experiments are welcome in the comments as my exploration into the world of AI-assisted education continues.
The AI experiment – Learning Objectives Edition
In this experiment to determine the best AI for Learning Objectives, we compare the capabilities of three AI tools: ChatGPT, Bard, and Bing, for generating learning objectives. (all AI are used in their free version)
We use a simple prompt to create Learning Objectives for an online course about on-page SEO. Then, we evaluate the outputs using a matrix including relevant criteria.
Prompt: Create the learning objectives of an online course about on-page SEO.
In order to evaluate the outputs, I figured out a matrix on how to grade the AI outputs and make them comparable.
Don’t miss out on Which AI is Best for Generating Learning Objectives?
Learning Objectives evaluation matrix
The evaluation matrix I’ve carefully crafted for this post is our trusty guide to dissecting and assessing with precision. This matrix ensures that we leave no stone unturned. It’s our tool for systematically breaking down the nuances of each output, allowing valuable insights and an informed perspective.
Here is a possible matrix with criteria to evaluate learning objectives in industrial design, based on the information I found from the web search results.
Grading Matrix
| Criteria | Description | Assessment Criteria |
|---|---|---|
| Alignment | The learning objectives should be aligned with the business goals, the performance gaps, and the learners’ needs and expectations. | The learning objective of designing a sustainable product is aligned with the business goal of reducing environmental impact, the performance gap of low customer satisfaction, and the learners’ need of enhancing their creativity and problem-solving skills. |
| Specificity | The learning objectives should be specific, clear, and concise, stating what the learners will be able to do or demonstrate after the learning intervention. | The learning objective of applying the principles of ergonomics to improve user comfort and safety is specific, clear, and concise, stating what the learners will be able to do after the learning intervention. |
| Measurability | The learning objectives should be measurable, using observable and quantifiable indicators of learning outcomes. | The learning objective of creating a prototype of a new product using CAD software is measurable, using observable and quantifiable indicators such as accuracy, functionality, and aesthetics. |
| Achievability | The learning objectives should be achievable, realistic, and attainable for the learners within the given time frame and resources. | The learning objective of developing a portfolio of industrial design projects is achievable, realistic, and attainable for the learners within the given time frame and resources. |
| Relevance | The learning objectives should be relevant, meaningful, and applicable to the learners’ current or future work context. | The learning objective of evaluating the usability and user experience of a product is relevant, meaningful, and applicable to the learners’ current or future work context. |
| Outcome-based | The learning objectives should be outcome-based, focusing on the desired results or changes in behavior rather than the activities or tasks involved in the learning process. | The learning objective of improving customer satisfaction and loyalty by designing user-friendly products is outcome-based, focusing on the desired result rather than the activity. |
Resources to create the matrix
I used the following sources to create the matrix:
- Learning evaluation, impact and transfer | Factsheets | CIPD
- Learning needs analysis | Factsheets | CIPD
- Course Objectives & Learning Outcomes | Course Design | Teaching Guides …
- 4 Learning Evaluation Models You Can Use – eLearning Industry
Rubric for assessing the AI output
I will use a simple and generic 1-5 grading rubric, to not overcomplicate things.
- Very poor
- Poor
- Good
- Excellent
- Outstanding
Now, after all this prep, let’s get started with the Learning Objectives! I will stick to the same on-page SEO course, which we used in the other AI challenges!
Which AI is Best for Generating Course Outlines?
Output per AI – Learning Objectives
I did not modify the output of the AI tools. Find the three answers from Bard, ChatGPT and Bing below. And as you will see, there are not only 3 answers!
Here is the prompt I used for all queries to generate learning objectives. This prompt is simple. Stay tuned because in the future we will work with more elaborate prompts!
Prompt: Create the Learning Objectives of an online course about on-page SEO.
Bard
Here are some learning objectives for an online course about on-page SEO:
- Define on-page SEO and its importance.
- Identify the key factors of on-page SEO.
- Apply on-page SEO best practices to your website.
- Optimize your website’s content for search engines.
- Use keywords effectively throughout your website.
- Improve your website’s structure and navigation.
- Optimize your website’s images and videos.
- Build internal links to improve your website’s architecture.
- Use external links to cite credible sources and improve your website’s authority.
- Monitor your website’s SEO performance and make adjustments as needed.
Here are some more specific learning objectives that you could include in your course:
- How to conduct keyword research to identify the right keywords to target for your website.
- How to write titles and meta descriptions that are optimized for search engines.
- How to use header tags and subheadings to structure your content.
- How to optimize your images for search engines by using descriptive file names and alt text.
- How to build internal links to improve your website’s architecture.
- How to use external links to cite credible sources and improve your website’s authority.
- How to use Google Analytics to track your website’s SEO performance.
You can tailor these learning objectives to the specific needs of your audience. For example, if you are teaching a course for beginners, you might focus on the more basic aspects of on-page SEO, such as keyword research and title tag optimization. If you are teaching a course for more experienced users, you might focus on more advanced topics, such as internal linking and external linking.
No matter what your target audience is, be sure to make your learning objectives specific, measurable, achievable, relevant, and time-bound. This will help you to develop a course that is effective and engaging.
Well, no comment Bard. You not only failed this simple task, but you also messed up the formatting of this paragraph. Is the list text all in bold?!
It is common sense that learning objectives should always follow one line:
By the end of this course….
Chat GPT
Surprising output with ChatGPT too. Instead of putting the standard By the end of this course…. at the beginning, ChatGPT inserted this partially in the learning objective.
In addition, the first objective, ‘Understanding the Fundamentals’ does not sound convincingly measurable, a key criterion for learning objectives, especially for a technical topic.
Well, that didn’t turn out as expected. Bard and ChatGPT put out some text that does not resemble learning objectives at all! The format was unclear, and the objectives were somehow vague.
Bing –
Bing nailed it immediately at the first attempt. Well done! It uses the standard layout for course objectives and includes measurable outcomes using Bloom’s taxonomy.
This makes Bing already the clear winner!
But I am curious now. Let’s try out a different prompt for Bard and ChatGPT.
Prompt #2
Prompt: Define the Learning Objectives of an online course about on-page SEO.
Hard to believe, but Bard failed again and produced a similar output. Weird, because I generated learning objectives with Bard in the past and never realized it was doing it wrong with a simple prompt! I will not even include this embarrassment here.
At least ChatGPT produced a nice outcome now on the second try.
ChatGPT (2nd try)
Certainly, here are defined learning objectives for an online course about on-page SEO:
Learning Objectives: On-Page SEO Course
By the end of this course, participants should be able to:
- Understand On-Page SEO Fundamentals: Explain the core concepts of on-page SEO, including its role in search engine optimization and its impact on website visibility.
- Conduct Effective Keyword Research: Demonstrate the ability to conduct comprehensive keyword research, select relevant keywords, and identify keyword optimization opportunities.
- Create SEO-Optimized Content: Develop high-quality, SEO-friendly content that aligns with target keywords, engages users, and meets industry standards.
- Optimize On-Page Elements: Optimize critical on-page elements, such as title tags, meta descriptions, headings, images, and URLs, to improve search engine rankings.
- Implement Technical SEO Practices: Apply technical SEO techniques to enhance website performance, including page speed optimization, mobile-friendliness, and structured data implementation.
- Utilize SEO Tools: Proficiently use popular on-page SEO tools and analytics platforms to monitor and improve website performance.
- Adhere to SEO Best Practices: Adhere to industry best practices for on-page SEO, staying updated with emerging trends and innovations in the field.
- Assess and Optimize: Evaluate on-page SEO efforts through assessments and analytics, identifying areas for improvement and implementing optimization strategies.
These defined learning objectives provide a clear roadmap for participants in the on-page SEO course, ensuring they acquire the knowledge and skills needed to excel in this field.
Well done on the second attempt ChatGPT! Clear objectives with measurable outcomes and action words. Nice.
Well, let’s take care of Bard, he seems to not align this simple prompt to learning objectives’ best practices.
So I will try again, giving it a hint with the following prompt.
Prompt #3
Prompt: Define measurable Learning Objectives of an online course about on-page SEO.
Bard’s 3rd attempt
Here are some measurable learning objectives for an online course about on-page SEO:
- By the end of the course, students will be able to:
- Define on-page SEO and its importance, and explain how it can help improve their website’s ranking in search engine results pages (SERPs).
- Identify and implement the key factors of on-page SEO, such as keyword optimization, title tag optimization, header tag optimization, image optimization, and internal linking.
- Use Google Analytics to track their website’s SEO performance and identify areas for improvement.
Here are some specific examples of measurable learning objectives:
- By the end of the course, students will be able to:
- Identify 10 relevant keywords for their website using a keyword research tool.
- Write a title tag and meta description for their website that are optimized for their target keywords.
- Optimize the header tags and subheadings on their website to improve its structure and readability.
- Optimize the images on their website by using descriptive file names and alt text.
- Build 10 internal links on their website to improve its architecture.
- Use a tool like Google Analytics to track their website’s SEO performance metrics, such as organic traffic, click-through rate (CTR), and bounce rate.
These learning objectives are measurable because they specify what students will be able to do or achieve by the end of the course. For example, students will be able to identify 10 relevant keywords for their website, write a title tag and meta description that are optimized for their target keywords, and optimize the header tags and subheadings on their website to improve its structure and readability.
You can use these learning objectives to develop assessments for your course. For example, you could give students a quiz to test their knowledge of on-page SEO principles, or you could ask them to complete an exercise in which they optimize a web page for search engines.
By using measurable learning objectives, you can ensure that your course is effective and that your students are learning what they need to know.
Bard, finally passing the finishing line! And it is a nice result, showing the core learning objectives and some optional ones. In addition, Bard always includes some nice hints and additional features.
But to sum it up, failing the test twice will result in losing some points in the final assessment.
In the following chapter, we let ChatGPT evaluate the best AI for learning objectives.
ChatGPT evaluates the Learning Objectives
Based on the provided matrix and grading criteria, here’s an evaluation of the learning objectives generated by ChatGPT for Bard, ChatGPT, and Bing for an online course about on-page SEO:
Bard:
- Alignment (2/5): Bard’s initial attempt did not align with the standard format of learning objectives. In the second attempt, it improved alignment but still included additional text, affecting clarity.
- Specificity (2/5): Bard’s initial attempt lacked specificity and measurable outcomes. The second attempt improved, but some objectives could still be more specific.
- Measurability (3/5): Bard’s objectives in the second attempt were measurable, but the first attempt lacked measurable outcomes.
- Achievability (3/5): The objectives in Bard’s second attempt were realistic, but the first attempt did not specify achievability.
- Relevance (3/5): Both attempts were relevant to the topic but lacked clarity in the first attempt.
- Outcome-based (2/5): Bard’s objectives in the second attempt were somewhat outcome-based, but the first attempt did not focus on desired results.
ChatGPT:
- Alignment (3/5): ChatGPT’s initial attempt lacked alignment with standard learning objective formatting, but the second attempt improved alignment.
- Specificity (3/5): The objectives in ChatGPT’s initial attempt were not entirely specific. The second attempt improved but still had room for more specificity.
- Measurability (3/5): ChatGPT’s objectives in the second attempt were somewhat measurable, but the first attempt lacked clear measurable outcomes.
- Achievability (3/5): Both attempts by ChatGPT did not specify achievability.
- Relevance (3/5): ChatGPT’s objectives were relevant to the topic in both attempts but needed more clarity.
- Outcome-based (2/5): ChatGPT’s objectives in the second attempt were somewhat outcome-based, but the first attempt did not focus on desired results.
Bing:
- Alignment (4/5): Bing’s objectives were well-aligned with the standard format of learning objectives.
- Specificity (4/5): Bing’s objectives were specific and clear, with measurable outcomes.
- Measurability (4/5): Bing’s objectives were measurable, with observable and quantifiable indicators.
- Achievability (4/5): Bing’s objectives were realistic and attainable within the given context.
- Relevance (4/5): Bing’s objectives were relevant and applicable to the topic.
- Outcome-based (4/5): Bing’s objectives were outcome-based, focusing on desired results.
Overall Assessment:
- Bard’s initial attempts were subpar, affecting the overall assessment. The final attempt improved significantly but still lacked in certain aspects.
- ChatGPT’s second attempt was better than the initial one but had room for improvement in specificity and measurability.
- Bing excelled in providing well-structured, clear, and measurable learning objectives aligned with standard formatting and best practices.
In summary, Bing demonstrated the best performance in generating learning objectives for an online course about on-page SEO, while Bard and ChatGPT had some challenges in the initial attempts but improved in subsequent tries.
| Criteria | Bard | ChatGPT | Bing |
| Alignment | 2 | 3 | 4 |
| Specificity | 2 | 3 | 4 |
| Measurability | 3 | 3 | 4 |
| Achievability | 3 | 3 | 4 |
| Relevance | 3 | 3 | 4 |
| Outcome-based | 2 | 2 | 4 |
| Total Points (max. 30) | 15 | 17 | 20 |
Conclusion: What is the best AI for Learning Objectives?
In hindsight, I didn’t expect this simple experiment about the best AI for learning objectives to become so interesting. There can only be one clear winner in this competition, and that is Bing! Its opponents failed to produce acceptable results in the first prompt.
Obviously, I was very rigorous in not prompting again. With slight adjustments to the prompt, all three AIs produced very good results. The outcomes are measurable, clear and can easily be aligned with assignments.
I was surprised that the three generated learning objectives are very different in many aspects. To be honest I expected similar outcomes and only minor differences.
What did I learn? First of all, for my use case in particular, I want ideas on how to manually design the learning objectives for my courses. With all the variety in both Bloom’s Taxonomy and the different perspectives on the topic, I most likely use Bard, ChatGPT and Bing together. A medley of all three, manufactured in my personal style, will bring the best outcome.
What is your best AI for Learning Objectives? I am glad to read more about your approach. Do you like my AI experiments?
Leave your feedback and ideas for future AI experiments in the comments below!
