YT YouTube2Text

YouTube2Text → Transcripts → Kapitel 2 Statusdiagnostik oder Verlaufsdiagnostik

Kapitel 2 Statusdiagnostik oder Verlaufsdiagnostik — Transcript

by Markus Gebhardt · 2,480 words · 387 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00Hi and hello. Welcome to the second
0:03part of this lecture series. Today
0:06we'll look at the question of status
0:08and progress diagnostics, and which
0:10diagnostic models we use or are
0:12generally familiar with. The first
0:14question is, what is status diagnostics
0:17, or why do I call it status
0:19diagnostics? Well, it’s simply about
0:22a profile, a status at a specific point
0:24in time. And in most classical
0:28diagnostic processes, where you go to a
0:31school psychologist or, in some form,
0:33want to know everything about this or
0:35that problem at one measurement point,
0:37you likely use various test instruments
0:39to clarify several aspects to decide:
0:42Is there a need for support, and if so,
0:44what does it look like? And whenever I
0:49need a profile, so to speak, where I
0:52clarify multiple aspects to then decide
0:54: Yes, this is exactly where support
0:56should begin or here in. This structure
1:00should be changed, then I need
1:01comprehensive knowledge, and whenever I
1:04need comprehensive knowledge, I also
1:06need more time, and that's when status
1:08diagnostics is the answer. On the other
1:13hand, if I already know it’s about a
1:15math difficulty, about this or that
1:17specific area, and I want to reflect on
1:20support decisions and discuss whether
1:22the support is working or not, then
1:24progress diagnostics is the better
1:26choice. Then it's not about observing
1:32many things simultaneously, but rather
1:34the relevant aspects of the support.
1:39For example, if I want to check
1:41arithmetic fluency in addition up to
1:43100, or just determine: can they handle
1:46the tens transition or not? If I have a
1:52yes or no question, I can answer it
1:54easily by giving 20 tasks. And if I see
1:58they can solve most of them and they
1:59have the right strategy, then that’s
2:01wonderful. However, if it involves
2:08longer-term support—meaning we are
2:10working on addition up to 100 over a
2:12longer period, and we’ve considered
2:14several interventions or actions, and
2:16only slow progress is visible, which is
2:19usually the case with learning
2:20difficulties. Then the idea of a
2:26learning progress diagnostic graph is
2:29very good, because then I can see
2:30exactly which support is working and
2:32what is slowly improving. Then you can
2:35perhaps subdivide that, and I basically
2:38have the idea of testing multiple times
2:40. So, if you look at the individual
2:42case examples or the evaluation below,
2:45you could, for instance, do a pre-, mid
2:47-, and post-phase. And the idea of
2:51measuring more frequently is simply
2:53that I can more clearly determine where
2:55the true value lies. Because we always
2:59have various everyday factors and
3:01unsystematic errors that influence us,
3:03and on the other hand, we also have
3:05support or a support phase where we
3:07want to see—perhaps with a bit of a
3:09delay, since people need to sleep on it
3:11—active learning. So just because
3:16you've briefly explained something
3:18doesn't mean it’s being implemented
3:20at the behavioral level, and hopefully,
3:22we then see a learning achievement, an
3:24increase, simply a higher number of
3:26tasks solved. And that is exactly the
3:29idea behind formative assessment. This
3:31means that during the support phase, I
3:33already get results and see whether the
3:35support is working or not. That is, of
3:37course, only one aspect when I am
3:39practicing and providing support. I
3:41naturally also see how the person
3:43reacts, how they act, what feedback
3:45they want, and what strategies they
3:47choose to solve this or that task. That
3:52is also a second important aspect of
3:54support, but here in terms of
3:55diagnostics, one can simply look at one
3:58thing, and that is represented by this
4:00graph. That is the core idea of
4:03progress monitoring. How does this
4:05relate to diagnostic models in
4:07themselves? Here there was, on one hand
4:10, the classic, traditional idea of
4:13special education diagnostics, that we
4:15have a very strong integration of
4:17diagnostics, support, and evaluation.
4:20So, classically speaking, there was
4:22actually only diagnostics, and the
4:23support was something else entirely and
4:25sometimes had nothing to do with each
4:27other. But that was never good. And
4:30that is why the interplay has always
4:32been an important aspect, and Mr.
4:34Heimlich, along with other special
4:36educators, has captured this in various
4:38models. And in this process, you simply
4:42conduct status diagnostics twice—once
4:44at the beginning and once at the end—
4:46with an intervention in between. And
4:50what is important is that the support
4:52is relatively specifically linked to
4:54the diagnostics, and the evaluation
4:56aims at a summative result exactly at
4:58the end of the support: Did the support
5:00plan hold up or not? And if that were
5:06practically implemented, we as
5:08scientists would be in agreement on
5:10everything regarding the practical
5:12progress or the critique of support
5:13plans, especially in the field of
5:15special education. If we look at
5:18various studies from Mose Opz. Then we
5:21see: status diagnostics are sometimes
5:24done. Support plans, and especially
5:26their evaluation, are rather not done,
5:29or at least they are not
5:30domain-specific or precise. The goals
5:33are also not clearly defined, and the
5:35evaluation does not take place to the
5:38extent one would wish; rather, it is
5:40usually based on observations or
5:42smaller elements. But if you consider
5:45what success in a pedagogical measure
5:47actually is, it is simply that I reach
5:49the goal. So, in terms of
5:53output-oriented management, I should
5:55really consider whether I am reaching
5:57my goal in the end, whether I am
5:59effective, and see that as the
6:00essential element of my pedagogy,
6:02rather than giving special weight to
6:04any other aspects. The model by Werneu
6:08is essentially similar and also places
6:10process diagnostics in the foreground
6:12here. It also has this three-step
6:15approach, so it is not different in any
6:17way. I just spent a long time searching
6:19to see if I could find any more models
6:21or people who have formulated it this
6:23way. And that is where I found the two
6:24of them. In the meantime, there is a
6:28third model, the Gritt model, which, as
6:30far as I have understood it so far,
6:32focuses primarily on status diagnostics
6:35and various forms of support, and is
6:37more of a grid, similar to the grid
6:39developed by Ms. Lutz, which
6:41practically covers support planning or
6:43a support report. But I am sure this
6:47model will be developed further, and
6:48then I will include it as well. And if
6:51you know of any other models, please
6:52feel free to write them in the comments
6:54. Many models that are linked to
6:57diagnostics and support are more
6:58subject-didactically oriented, meaning
7:01based on a concrete subject. I have not
7:04included those here and am therefore
7:06not currently referring to them. But
7:08they are just as good, so to speak. As
7:11a second model or as an extension of
7:13this previous summative model with a
7:15beginning and an end, there is the idea
7:18of learning progress diagnostics. Mr.
7:22Helbert from Potsdam was the first to
7:24present it this way. I have expanded it
7:27a bit, and in case you are wondering
7:29why on earth it is green. That is
7:32because a mishap simply occurred during
7:34the formatting. What is important is
7:37that you can read it. And the idea is
7:40now similar to the model before. This
7:43is practically shown on the left now.
7:45You have an occasion, you have a status
7:47diagnosis, you determine the need for
7:49support, and you set the support goal.
7:53You probably do that once, or do one
7:55larger diagnostic at the beginning, but
7:57then you actually move into a kind of
7:59classroom-based assessment. You know
8:03roughly where the support goals are.
8:06They do not change significantly
8:08anymore; rather, it is much more
8:10important to reach the individual
8:11target areas of the support diagnostics
8:14. And when you are in the second step,
8:16it is actually more sensible to proceed
8:19formatively, because then you already
8:21know what the support is supposed to
8:23achieve, and the fundamental question
8:25is: is the support highly effective?
8:28That would be desirable, then
8:30everything works wonderfully. Then you
8:32can set the next support goal at the
8:34next level, so to speak. On the other
8:39hand, if the intervention isn't working
8:41, I should ideally find out immediately
8:43, perhaps through progress monitoring
8:45or formative assessment, because the
8:47moment it’s formative, I can adjust
8:49the intervention accordingly during the
8:52phase. And if we know anything in
8:56pedagogy, it's that feedback is always
8:58most effective when it is immediate.
9:02Most feedback, like when you take a
9:04test last week and get the results two
9:07weeks later, simply lacks stringency.
9:11Then you don't know where to start or
9:13how to improve the whole thing. So, the
9:15more immediate the feedback you receive
9:17and the sooner you can adapt, the
9:19better the whole process is. And it's
9:23the same with interventions: when you
9:25adapt them and say, "Ah yes, after the
9:27second session, as you plan further
9:29sessions, I reflect on: What's working,
9:32what isn't, what should I adjust, what
9:34other structure do I have, where could
9:36I tweak things?" These are exactly the
9:40decisions you need, and you don't want
9:42to wait until you have a result again
9:44after half a year. And that is exactly
9:49what progress monitoring is for—so
9:51you have immediate decisions—and that
9:53is becoming easier with computer-based
9:55diagnostics or certain learning games,
9:57and I think that’s where the future
9:59lies. Whether it's Leifumi or other
10:06things like learning lines or similar
10:07instruments, or now AI-based systems
10:09that perhaps also have learning games
10:11and interventions at the same time,
10:13regardless, there will always be some
10:15kind of feedback for the teacher. When
10:20do I need to make which decision, and
10:22the more immediate it is, the better it
10:24is. Now I am on chapter 2.3. This is
10:27theoretically a small leap back into
10:30history. Why am I starting again with
10:32the zone of proximal development? And
10:34Vygotsky, ultimately, is clear; many
10:37things in special education are built
10:39upon these ideas from the
10:41cultural-historical school. I am
10:44evaluating this entire script in terms
10:46of diagnostics, and this idea of the
10:48zone of proximal development is quite
10:50interesting because if you imagine
10:52taking current school achievement tests
10:54, you do them without help. So you
10:57don't get any support at all. So all
11:00ideas of testing, and even identifying
11:03a disorder, are based on current
11:05development, even though we as
11:07educators are more focused on the idea
11:09of future development, i.e., the zone
11:11of proximal development. Then the
11:14question is: is our testing system
11:16actually completely outdated, since we
11:18want to move toward development? Why,
11:21why can't we test in a way that brings
11:24us much closer to actual development? I
11:27mean, why must we always only capture
11:29the current status? That is exactly the
11:32question that Good and other
11:34psychologists asked themselves, leading
11:36them to the conclusion that we really
11:38need to adapt and change this. And the
11:44first or simplest idea one might have,
11:46which is in that sense a so-called
11:48dynamic testing, is simply to have
11:50multiple testing intervals. And, above
11:55all, Marco Hessels discovered that if I
11:57have an individual with an intellectual
11:59disability—someone who finds certain
12:02tasks difficult and for whom
12:03instructions are not always clear. And
12:06even though these individuals are
12:08tested quite frequently with IQ tests,
12:10it is not at all clear whether they
12:11actually understand all the
12:13instructions involved. And the CFT also
12:15has a few more difficult elements in
12:17its subtests, as I would describe them.
12:21And that is when the idea came up. Well
12:23, one could simply take the first test,
12:26discard it, and then proceed with the
12:28second test. And the second test is
12:32simply much more reliable because they
12:34have already become accustomed to the
12:36testing procedure and the environment,
12:38allowing them to truly reach their
12:40maximum potential in the test. That
12:45would be, so to speak, a first step.
12:49But dynamic testing actually goes much
12:52further, namely: The idea behind it is
12:54to truly enter the zone of proximal
12:56development, much like a teacher or
12:58mentor who helps one progress to the
13:00next level with the right guidance and
13:02helps one understand more difficult
13:04tasks. One could approach dynamic
13:13testing in the same way, by focusing
13:15not on the tasks and their difficulty,
13:17but rather on the number of prompts the
13:19person receives to understand the next
13:21task. And the fewer prompts someone
13:28needs, the higher their potential, the
13:30faster that person understands, and the
13:32faster they can reach the next level.
13:37And if you now imagine that we might
13:39have some kind of computer-based
13:41environment or learning games in the
13:43future, and in these games—however
13:45they are designed—they not only have
13:47current tasks but also provide hints
13:49and support, then we are already quite
13:51close to this idea of dynamic testing.
13:57And that is exactly the reason why I am
13:59including this in these psychology
14:01topics, so one simply considers: well,
14:03there are these current tests with
14:05their tasks at the current level, and
14:07there is a kind of competence track or
14:09learning ladder where I assume a higher
14:11level and I provide hints. And that is
14:16precisely where we arrive at this idea
14:18of the zone of proximal development.
14:22And the question arises much more about
14:25which hints or support measures are
14:27needed, and if so, how many? And those
14:32can just as easily be evaluated for
14:34diagnostics. 2.4 is data-based decision
14:38making. This is ultimately similar to
14:41the previous models with learning
14:43progress diagnostics or status
14:45diagnostics, in that it is about
14:46collecting information. Only the idea
14:50here is the question: what is relevant
14:52information, what is less relevant
14:54information, what is one-time
14:56information, and what is process
14:58information? So it is practically a
15:03model where I am aware of what
15:04information I am receiving, which
15:06information is truly important and
15:08relevant, and how I collect it further
15:10and discuss it in my team. And that is
15:15exactly what we actually need in school
15:17. That, firstly: everyone must be
15:19informed about what information is even
15:22available, what support decisions are
15:24associated with it, and how do we work
15:26together? And then you need a recurring
15:31workflow where I say: I first set a
15:33goal, then I collect the data and
15:35analyze it, then I take the measures,
15:37implement them, and evaluate them. And
15:41then I apply the evaluation again, and
15:44that also exists as a research cycle.
15:46Mr. Cool and I have already written
15:48papers on this, and many others have
15:50also contributed many ideas. And this
15:54is, so to speak, this model that Ms.
15:56Lutz has illustrated nicely here with a
15:58very good practical example, where she
16:01wrote a good article or contribution
16:03with a lot of practical relevance,
16:05where it is about the implementation of
16:07this model, and you can find it here. I
16:11will not go into detail on this article
16:13now.

About this transcript

This page contains the full transcript of Kapitel 2 Statusdiagnostik oder Verlaufsdiagnostik by Markus Gebhardt, generated from the public captions YouTube serves with the video. The transcript has 2,480 words across 387 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.