Gary Ackerman @garyackerman@qoto.org

If one really wants to be "data-driven," they must adopt a science-like approach to their data. Few realize how science deals with data, however. I tell #education folks they cannot be data-driven without:

The supremacy of observation and logic.

Avoidance of truth. Rather than demonstrating an idea is true, scientists demonstrate an idea is a false.

Accounting for variables. In education, we assume our interventions are the only factors affecting what we are measuring. This is just one example of fields where relevant factors are ignored.

Verification of observations. If others do not observe the same thing, you are probably wrong.

Peer review is necessary to evaluate one's work and ensure that necessary data was collected and that it was properly and logically analyzed and that observation supports conclusions.

Once a new paradigm has become established, there are typically some individuals who continue to work according to the old paradigm, but those individuals find themselves increasingly marginalized.

We have not seen a new #eduction paradigm in generations.

Earlier this year, I released Technology in Schools: Its Not Like this in Business under CC-BY-NC.

"My purpose in writing this book is to give readers a view into the work of managing information technology in schools. IT professionals will notice differences (some nuanced and some significant) between the needs and expectations of IT users in business and IT in school."

You can find an .odt and .pdf versions on my web site:
https://hackscience.net/technology-in-schools/

Once you can differentiate red herrings from meaningful practices, you will be very successful… but you will anger leaders.

If your course does not include some activities in which ethics of your field and its applications are the focus, it’s time to rewrite your syllabus.

Is it just the server I joined or are there an inordinate number of geoscience folks on Mastodon? I’m not composing, just seems a statistical anomaly.

Sitting outside in New England in November… it’s about 20 C (almost 70 F). Unseasonably warm. The geese are still flying south. They will not be fooled.

When you work as an educational technology professional, you as invited to join committees where “we are going to reinvent something.” What they really mean is “we want you to build a web site” and our only role is to email you thing to put on it and complain about the information that is on it.

I hang out with #educators who talk about how they are #data driven. They seem to misunderstand the importance of accurate interpretation of data.

While claiming to be "data-driven," educators are incredibly sloppy with their collection, analysis, management, and reporting of data. Their sloppiness derives from blind acceptance that tests measure what they proportion (and other unchallengeable assumptions), their reliance on single measures, and their lack of sophistication in analysis. Statisticians have created tools to help us understand large data sets. Let’s insist they start using those tools.

Two Types of Problems

Most jobs require workers to solve problems. Usually, we are taught to solve problems with this process: Set a goal, develop and implement a plan to solve it, and then decide if the goal was accomplished. That leads to new goals and the process is repeated.

1) One begins with goals-- we decide what we want to accomplish and how we will accomplish it.

2) Next, one decides what actions will result in the goal being accomplished. In designing these actions, planners assume that they know what will happen if they take the actions they developed. All that needs to happen is that the actions be taken or implemented.

3) After the actions are taken, one checks to see if the goals were accomplished. If not, then the steps are taken to redesign the actions or to improve the implementation of the actions.

This model is built upon the assumption that one can accurately predict the effects of actions. This is sometimes described as cause and effects. Everything we observe has a single cause (or a small number of causes) and one can know those clearly. The objective-based problem solving works best for problems that are:

- definable-- We can clearly identify the cause of the problem

- understandable-- We know what will happen if it is solved.

- consensual-- Most everyone agrees that it should be solved.

Someone who has a stroke has a definable (there is a blood clot preventing blood from flowing to a part of the brain), understandable (that part of the brain without blood will die), and consensual (the patient and his family and the medical team all want to clear to blood clot). Scholars refer to such problems as tame. Tame does not mean that it is not complex or important, only that the course of action is known.

Another collection of problems, especially those in the social sciences and those that deal with social issues and problems, are very difficult to define and understand, and not everyone agrees what to do or even that the problem should be solved.

These problems are referred to as wicked problems, and because cause and effect are difficult to establish with these problems, solving them is difficult or even impossible.

Making people “smart” is a good example of a wicked problem. No one really knows what it means to be smart (and smart depends on the situation-- knowing about computers is not much help when you need a plumber!), we don’t know how to make people smart (what helps me learn may not help you) and we don’t even agree that we should make people smart.

The importance of goals and objectives in the process is a relatively recent idea to be added to planners’ and managers’ work; before the 1960’s most workers focused on making their work more efficient or “better” and not “what should my system do?” In this figure, we see a version of “outcomes-based action” that has been common in recent decades:

What we mean by "they learned it:"

In addition to having strong foundational knowledge (what we traditionally understand learning to be), we want those who have “learned it” to be able to use it critically; they should be able to judge the quality of their knowledge and the degree to which it will suffice in the current situation. We want those who have “learned it” to be able to use it creatively; they should be able to use what they have learned to craft something that did not exist previously. We want those who have “learned it” to be able to use it pragmatically; when faced with real-world situations and challenges, an educated person should be able to use it to solve the problem. We want those who have “learned it” to value it and value the world differently because of it.

I'm interest in #teaching #learning #communitycollege #highered #onlinelearning #science #math #education #computers

Richard Feynmann served on the commission that investigated the Challenger disaster in 1986, and he observed that much empirical evidence had been ignored and that decisions that led to the disaster had been made for political reasons. Feynman concluded his observations, which appeared in an appendix rather than in the full report, with the statement “For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled.” Progressive discourse requires the group to regard reality as Nature intends, and observation is necessary to expand the factual basis of progressive discourse.

I do appreciate the extra characters in Mastodon posts... I keep posts shorter than my blog, but a paragraph sure seems better than 280 characters. Of course, I'm a fan of article abstracts also.