2015年1月10日 星期六

CLA final report 0110 (2015)

我目前想到的…就是應該要找一些authentic text來作測試材料。因為測試的題目應該是要印出來,且字體大小要維持統一、清晰,所以最好的情況還是我把他們統一成一個檔案印出來作試卷。(我最好快一點,因為我得留夠多時間讀Syntax)

可以讓我這樣採用的適合材料,我想初中高三個等級都各擷取兩部份文件,也就是至少會有6段text。至於內容,我想先不選擇學術傾向的東西。至於版權和難度不好控制(?)的問題,我想先避開studio classroom之類的東西。(我把列印用的檔案放這裡 https://www.dropbox.com/s/iuc1qpdexzl6mu9/CLA%20%20final.docx?dl=0

Topic

A case study for teaching a computer programmer to read English newspaper: could it be possible to boost reading speed and enhance understanding with a semantic-less system – geometric grammar rules?

Introduction

For the past decades, teachers and students are led to believe that fluent reading heavily relies on semantic knowledge. But I have an assumption that instructors can gap their ability through a simple syntactic system that work with little help from semantic meaning - the geometric grammar rules. I found little research have been done to figure out how the mnemonic devices help people with lower reading ability in English.  

Literature Review

Wait till I incorporate those listed in the references

Research Question

Can g.g. rules help readers with limited reading proficiency enhance their reading speed and understanding on articles ? (Even when they are traditionally considered as slightly below the threshold of reading this article fluently)

In other words, can g.g. rules help people with limited semantic understanding read better and understand better with virtually pure syntactic rules.

  • Baseline of the test: people need to know at least the differences between verb an preps.
  • Implication of the theory: people get better overall understanding and processing ability when they learned a semantic-less system like g.g. rules.

Methods and Methodology

Main ideas in designing the methods:

  • Use graphical items to represent the syntactic relation among words in English. (avoiding the influence from known or unknown word meanings.)
  • As participants acquired g.g. rules, test their fluency on using it.

Timed test:

  • Test them on receptive way (apply the rules to circle the text, then check the accuracy)
  • Test them on productive way in composing a short composition (self-introduction with guided questions.)

Methods I designed in this experiment. (There are four major steps in total)

Steps for choosing participants:

  1. Find at least 3 or 4 participants for this experiment, closely examine their progress. All of them will be receiving g.g rules training. I have very limited time resources so I have to utilize all the suitable participants I could find. It’s impossible for me to go back to PCSH at the end of the semester for this experiment. They are busy with finals exams at this moment.
  2. Since I really have very limited time to conduct the experiment, I could only test it on my girlfriend Sunya, a 28 years old computer programmer. But it proves later that this is a very good choice. She could provide me many useful information and made my future test more detailed.
  3. Total test time (including our discussion and her reflection): about an hour.

Steps for teaching them G.G. rules within very limited time (10 mins):

  1. (After work or at break to take care of my mom.) Explain the rule chart to them, and ask them to practice right away under my supervision. I’ve got suitable text for them. They should use text from elementary level,  such as Let’s Talk in English and partially from the easiest long text in Studio Classroom. (We are NOT choosing dialogues for the experiment!)
  2. Make sure they have mastered G.G rules.
  3. Make sure they do not derive the classes of the words “only” from semantic meanings. They need to understand the core of G.G. rules is to work without semantic meanings.

Steps for conducting the tests:

  1. I got audio recording during training, explaining and testing. The files could be found on my server. http://kmimayo.no-ip.org/mod/forum/discuss.php?d=8 
  2. noting down some important conversation in the recordings. This would help me clarify the actual problems they encountered when learning the system with little or no semantic clues. (I had done part of these in my following discussion)
  3. the chosen texts would be better if they are similar in syntactic forms. Their content is not that important. But it shall cause no fear for the participants.
  4. The post-test shall be slightly harder than the pre-test in syntactic complexity. (But it’s hard to measure that and I think I lost control of it in this experiment. You may find the text here https://www.dropbox.com/s/iuc1qpdexzl6mu9/CLA%20%20final.docx?dl=0 The first part of the Robin Williams is my pre-test. The second part is the post-test.
  5. The materials are collected and clipped from NYT, CNN, and Advanced Studio Classroom, Studio Classroom. I can not find suitable materials from Let’s Talk in English because most of them are dialogues.
  6. We found that the materials (from Advanced Studio Classroom) are edited from LA times. This make the text less authentic and difficult to read. I believe the syntactic form and complexity have been changed dramatically for the purpose of targeted readers in Advanced Studio Classroom. This is not very ideal for this experiment. But we found it after we finished the experiment.

Steps for analyzing the data of the experiments

  1. This is a timed test. We need a timer to measure how fast a participants can finish pharsing the provided text. (eg. 50 seconds in total to finish pharsing and circling the paragraph)
  2. Measure their finished time. (before and after g.g rules training)
  3. Find out their accuracy in applying g.g.rules. (this is not seriously measured here.)
  4. See if they have improved in reading speed. (at least in circling the words) => results: NO. 
  5. No stroop tests at this moment/stage

Data Analysis

Finished time of the tests:

  • Pre-test (1:43)
  • Post-test (2:44)
  • Mid-test (20min)

Clearly we could see it took Sunya more time to finish the post-test. This is contradictory to my hypothesis. We had a discussion about the phenomena. Sunya even found that she’d be influenced by the unfamiliar rules when she just started the post-test. For anyone who wants to listen to the six recordings, please visit my server http://kmimayo.no-ip.org/mod/forum/discuss.php?d=8 

Here are 6 recordings from our discussion (I will find ways to imbedded them into the blog you’re viewing. But please move to my sever for playing them directly from the web.)

From what we got in the timed test, we have some findings that are contradictory to our assumptions. They are worth mentioning:

  1. Why does it take longer to finish the post-test?
  2. What’s the differences before and after learning g.g rules?
  3. How long would it take to internalize the rules? (so that it could actually benefit the learners’ reading and understanding of the text that are slightly above their manageable level. )
  4. How to conduct a suitable post-test ?
  5. What’s the limitation of teaching and learning g.g rules?

The first question: why does it take longer to finish the post-test? Because the participant is not familiar with the new system. She’d be spending extra energy to meet the requirements rather than parsing the text rapidly. She needs to go back and made sure that the verbs she marked is correct.

My assumption on teaching g.g rules is only partially correct. I am right about the simple rules and ease of teaching them. It took me less than 5 minutes to teach her the rules and less than 5 minutes to practice it.

I was wrong on two simple points:

  1. I didn’t expect the delayed post-test, even delayed for only 2 minutes, would have such huge influence on her.
  2. I didn’t realize, or I had long forgotten that the time to internalize even the simplest rules is much much longer than 20 minutes.

Let us discuss them from the first one. Usually an immediately post-test or a delayed for 2 minutes post-test won’t make much differences. But here Sunya reported to me that she had lost her momentum to parse the post-test as she did in the mid-test. She needs to reengage herself to get the habit of finding the main verbs. (aka. follow exactly what the g.g rules tell you to do)

This is actually quite the same issues in the second point here. A newly learned rule can not possibly internalized within 20 minutes, no matter how simple it is. I had long forgotten that I took at least 6 months to internalize the system myself. That is also a time when I first learning English and its grammar formally. I was 13 years old then. These conditions somehow intertwine with other factors and stably became my internal behavior for parsing English text. I have synesthesia ever since then, and I believe it helps me with parsing text. I have wondered if anyone could build the neural co-relation like I did in my early stage of learning the language. But it may take an adult longer time to reach the same effect even when they have better meta-linguistic ability. I have some suggestions for future researchers who are interested in synethesia and geometric grammar rules: please bear in mind that synethesia could be trained, we just don’t know under what condition it would happen or last for long. Can adults acquire synethesia as easy as children did? Does their meta-linguistic ability hinder their ability to perceive patterns like syntax in another intuitive way? If so, how could we apply g.g rules for those in need of help. Those who suffered from poor efficiency on reading long and difficult text, or those who have dyslexia? Some of them are not children in classroom, but engineers or professionals in all walks of life who need to study a lot of documents everyday.

 

Limitations and Conclusions

Sunya and I have some discussion on g.g rules. She immediately captured the core concepts of it and help me clarify some ambiguities in my design on the test.

In general, g.g rules are only helpful for learners with certain level of proficiency. They need to know basic English grammar and they shall be able to differentiate at least verbs and other parts of speech. The minimal requirements of g.g rules is not that high, in my respect. But Sunya points out the learners need to know the differences between main and adjunt clauses. This will largely influence their performance on g.g rules test.

Test like this can not measure reading speed directly, nor can it guaranties that the testers are following exactly what g.g rules tell them to do even when they can parse the text faster. 

We overcame some of these dilemmas by setting the necessary standards to meet the minimal requirements of g.g rules:

  1. ONLY circle the main verbs, including the be-Verbs.
  2. For your convenience, you may underline other parts of speech, such as V-ing or Vpp.
  3. Measure the tester not only on their finished time, but also rating on their understanding of the provided text.
  4. If possible, interview the testers after the test.
  5. Tell them to meet the balance point between understanding of context and structural parsing speed. Don’t just parse and circle.

We don’t need the square parts since one can easily tell the subjects lie right before the main verbs. For an advanced user like me, I could even circle some of the main verbs rather than all of them. This would make my parsing speed even faster.

We hope ERP devices may be added for future study. The next step would be fooling intermediate-level g.g rules users with stroop tests. Finally we hope to see the fMRI test on an advanced user parsing the text. See if he really has formed synesthesia in his brain, and prove that synesthesia could be acquired and obtained even after puberty. This would be a good news for those who suffer from dyslexia and poor reading efficiency.

This is a smaller part of my research on incorporating Gestalt theory into teaching vocabulary. Please visit to see more related research. http://fall-cicada.blogspot.tw/2014/12/vocabulary-teaching-data-analysis.html 

Anyone who is interested in these topics are welcomed to contact me.

 

References

  1. Nilsen, A., & Nilsen, D. F. (2003). A new spin on teaching vocabulary: A source-based approach. Reading Teacher, 56(5), 436-439.
  2. CARPENTER, K. k., GEHSMANN, K. k., SMITH, R. r., BEAR, D. b., & TEMPLETON, S. w. (2009). Learning Together: Putting Word Study Instruction into Practice. California Reader, 42(3), 4-18.
  3. Sabeti, S. (2012). Reading graphic novels in school: texts, contexts and the interpretive work of critical reading. Pedagogy, Culture & Society, 20(2), 191-210. doi:10.1080/14681366.2012.672336
  4. Ho-hui, L. (2012). Teaching Vocabulary Through K.K. and Dictionary Phonetic Symbols to Taiwanese Elementary Students: A Collaborative Action Research Approach. English Teaching & Learning, 36(2), 35-78. doi:10.6330/ETL.2012.36.2.02
  5. Khabiri, M., & Rezagholizadeh, R. (2014). The Comparative Effect of Teaching Spelling through Mnemonics and Phonics on EFL Learners' Vocabulary Achievement and Retention. Theory & Practice In Language Studies, 4(7), 1463-1468. doi:10.4304/tpls.4.7.1463-1468
  6. Jarvis, L. H., Merriman, W. E., & Barnett, M. (2004). Input That Contradicts Young Children's Strategy for Mapping Novel Words Affects Their Phonological and Semantic Interpretation of Other Novel Words.Journal Of Speech, Language & Hearing Research, 47(2), 392-406. doi:10.1044/1092-4388(2004/032)
  7. Gill, S. R. (2006). Teaching rimes with shared reading. Reading Teacher, 60(2), 191-193. doi:10.1598/RT.60.2.9
  8. Thompson, G. B., McKay, M. F., Fletcher-Flinn, C. M., Connelly, V., Kaa, R. T., & Ewing, J. (2008). Do children who acquire word reading without explicit phonics employ compensatory learning? Issues of phonological recoding, lexical orthography, and fluency. Reading & Writing, 21(5), 505-537. doi:10.1007/s11145-007-9075-9
  9. George Lakoff

 

 

(以下放material相關的東西)

 

高級(NYT, BBC, CNN, LATimes)

NYT

[Search Continues for Woman Linked to Paris Attack Suspects - NYTimes.com](http://www.nytimes.com/2015/01/11/world/europe/charlie-hebdo-attack-paris-search.html?hp&action=click&pgtype=Homepage&module=b-lede-package-region&region=top-news&WT.nav=top-news&_r=0)

[Wrongly Convicted Man Was His Own Best Advocate - NYTimes.com](http://www.nytimes.com/2015/01/10/nyregion/wrongly-convicted-man-was-his-own-best-advocate.html)

[More Websites and Stores Rent Out Wedding Gowns - NYTimes.com](http://www.nytimes.com/2015/01/11/fashion/weddings/more-websites-and-stores-rent-out-wedding-gowns.html?_r=0)

[Cartoonists, killed for their art, and humanity - CNN.com](http://edition.cnn.com/2015/01/07/opinions/opinion-cartoonist-charlie-hebdo/index.html)

BBC News - State pension: 'Only 45%' to get full new payout
http://www.bbc.com/news/business-30777166

BBC News - In search of a personalised diet
http://www.bbc.com/news/magazine-30709297

 

 

CLA final

A look back at the talented actor and his movies

Robin Williams seized the day in every role

Shock came first on hearing the news that the master comic mind we knew as Robin Williams was dead at 63 [last August], apparently overwhelmed by the kind of despair that he lightened for so many of the rest of us.

With a mind forever moving at warp speed, conjuring up humor unlike anything we had seen before and will likely ever see again, Williams was a friend, the kind who made our world better. At his height, he electrified whatever entertainment he touched.

Though TV brought his breakthrough—I still remember his remarkable alien invasion in Mork & Mindy—film quickly became [his] main forum for talent.

Eyes twinkling with mischief made it clear that whatever else was going on, he had your back. He could be anything you wanted.

Williams’ first big-screen mark came in 1980, in the live-action Popeye, with Williams transforming himself with a pipe and outsize forearms.

Williams’ dramatic side

But it was in flexing his cerebral side in 1982’s The World According to Garp that Williams found his footing in film. This was no comic role. He surprised us. He could do drama; he could do more than tap that distinctive funny bone.

In the years after, Williams continued to surprise, making a habit of leavening his comedies with drama. His Russian defector’s last stand at Bloomingdale’s, in Paul Mazursky’s Moscow on the Hudson, is among his most endearing. His homeless dreamer in The Fisher King [is] among his sweetest and saddest.

Williams’ wisecracking wake-up call for soldiers in Good Morning, Vietnam brought his first Oscar nomination in 1988. A decade and two more nominations later—for The Fisher King and Dead Poets Society—he won after two unknown writers, Ben Affleck and Matt Damon, gave Williams the role of a lifetime.

Five Fun-filled Festivals - Winter festivals you won’t want to miss 你不會想錯過的冬季節慶

Most of us consider winter a great time to stay inside with a good book. Yet the winter months offer great outdoor events, including many fun and unique festivals.

The Ice Festival at Lake Khövsgöl, March 2, 3

To get to this festival, you’ll need to travel to “the ends of the earth!” The remote region located in Mongolia is the perfect setting for coldweather activities.

Lake Khövsgöl, called the Blue Pearl of Mongolia, freezes up to 1.5 meters deep. Frostcovered trees and mountains circle the lake, creating a winter playground.

Each year local residents have a two-day celebration. Wandering reindeer herders also join the fun, sharing their traditional songs and dances.

The activities take place on the frozen lake and include sled races, skating and much more. There’s an ice tower with a slide and an ice house with hot and cold drinks.

Visitors are always surprised to see how much weight the lake can hold. People, horses and even bonfires can be found on Lake Khövsgöl!

Cartoonists, killed for their art, and humanity

2011, masked gunmen broke the hands of Syrian cartoonist Ali Ferzat as a warning to stop him from satirizing Syrian President Bashar al-Assad. In 1987 Palestinian cartoonist Naji Al-Ali, creator of the character Handala, was killed outside his office in London.

Charlie Hebdo was sometimes rank with a particularly French form of xenophobia. But the bravery of Charbonnier and his staff in the face of death threats and firebombings was something to marvel at.

I mourn my fellow artists. I hope, perhaps futilely, that their murders will not be used to justify a backlash on the Muslims in France. Though I'm an atheist, I say a small prayer: for irreverence, for love, and for art.

BBC News - In search of a personalised diet

The study looked at three types of overeaters. Feasters who find it hard to stop eating once they start, constant cravers who feel hungry all of the time and emotional eaters who turn to food when they get stressed or anxious.

When it comes to feasters, research shows hormones play a big part in their eating habits. In particular, they produce low levels of certain gut hormones that are released when food arrives in the intestines. These chemical signals travel through the blood to the brain and tell the body when it has had enough food and should stop eating.

"Some people have astonishingly low levels of certain gut hormones and are not getting those signals," says Susan Jebb, a professor of diet and population at Oxford University.

Constant cravers always want to eat and their "hungry brains" often want fatty and sugary foods. Scientists know certain genes make people this hungry. They disrupt the way signals are sent to the brain telling it to stop eating, tricking it into thinking fat stores continually need replenishing.

"The role genes play in losing weight is unequivocal, but due to changes in technology we are beginning to find out what these genes are," says geneticist Dr Giles Yeo, from Cambridge University.

Emotional eaters reach for food when they are stressed or anxious. When the brain perceives a person is in difficulty, it triggers changes to the body, like the heart rate increasing. This stressed state makes it even harder for people to overcome what they are facing. This sort of eater has developed habits that are hard to break and in stressful times their brains seek out a reward.

"People often think diets are about willpower," says Jebb. "Forget that, diets are about habits. There has never been a study that says people can will themselves to lose weight, but they can change their habits."

More Websites and Stores Rent Out Wedding Gowns - NYTimes.com

“Men have been able to rent a tux, from Armani to Versace, for years,” said Jessica Berriman, a founder of Get the Gown, which opened last September and rents out, for an average $1,000 a week, dresses that typically cost $5,000 to $13,000. Among the labels on offer are Carolina Herrera, Oscar de la Renta and Vera Wang. Late returns cost $50 a day.

Her customers “don’t want to buy gowns they’re going to wear once,” Ms. Berriman said. “But they’re still very particular about what they want.”

With such discerning tastes driving them, some rentals may not sound like stand-alone bargains. But several shops emphasized that they catered just as much to trend-consciousness as practicality.

One Night Affair carries a mix of bridal fashions, from unlabeled samples to Badgley Mischka and Galia Lahav gowns. The dresses cost from $100 to $1,000 for five days, plus in-house tailoring costs. The shop also rents out custom-made dresses, starting at $299.