1
00:00:00,000 --> 00:00:05,880
This is the Reading Instruction Show. I'm your host as always Dr. Andy Johnson.

2
00:00:05,880 --> 00:00:14,040
Topic of today's podcast, How Science Works for Colon, Reading Science, Reading

3
00:00:14,040 --> 00:00:20,880
Research, and Reading Theories. So let's get into it. Just a bunch of theory you

4
00:00:20,880 --> 00:00:26,960
say. What often happens under the guise of reading science is that one study or

5
00:00:26,960 --> 00:00:34,920
one article or one report is used to prove things or to justify a policy

6
00:00:34,920 --> 00:00:42,680
program or practice. And this is not how science works, science of reading people.

7
00:00:42,680 --> 00:00:49,800
No single study should be the final word on anything. That would be a naive

8
00:00:49,800 --> 00:00:57,000
understanding of research in the social sciences. Instead, each research study

9
00:00:57,000 --> 00:01:03,680
creates a data dot and each data dot becomes part of a dot-to-dot picture,

10
00:01:03,680 --> 00:01:12,280
which is a theory. This is why theories are important. Now, a theory in science

11
00:01:12,280 --> 00:01:21,120
is different from an untested assumption. That would be a hypothesis. A theory is a

12
00:01:21,120 --> 00:01:29,400
way to explain a set of facts. Theories are used to understand phenomena. A

13
00:01:29,400 --> 00:01:34,440
theory is like a dot-to-dot picture connecting a series of research-based

14
00:01:34,440 --> 00:01:42,240
data dots. Theories are made up of many research-based data dots and different

15
00:01:42,240 --> 00:01:50,000
theories explain different data dots differently. Robust theories connect many

16
00:01:50,000 --> 00:01:59,080
data dots. Weak theories connect only a few data dots. Robust theories account for

17
00:01:59,080 --> 00:02:05,560
a lot of facts. Weak theories leave out important facts or have them unaccounted

18
00:02:05,560 --> 00:02:12,640
for. Now, there are different theories of reading, all of which are made up of

19
00:02:12,640 --> 00:02:18,800
data-dot research facts. Theories help us understand the process of reading and

20
00:02:18,800 --> 00:02:25,840
make good decisions about reading instruction. Using research-based theory

21
00:02:25,840 --> 00:02:32,960
to design programs and make decisions is appropriate. However, there's no such

22
00:02:32,960 --> 00:02:38,560
thing as correct theories and incorrect theories. As stated earlier, there are

23
00:02:38,560 --> 00:02:44,680
robust theories and weak theories. When theories are no longer useful in

24
00:02:44,680 --> 00:02:50,560
explaining a set of facts and understanding phenomena, they are

25
00:02:50,560 --> 00:02:58,000
replaced by other theories. Now, there are three views of reading. Let's start

26
00:02:58,000 --> 00:03:05,360
with the simple view of reading. This is a theory. Now, the science of reading is

27
00:03:05,360 --> 00:03:11,960
based on a weak theory called the simple view of reading. It's a weak theory

28
00:03:11,960 --> 00:03:17,880
because it doesn't account for a lot of data. This is sometimes called a bottom-up

29
00:03:17,880 --> 00:03:23,800
theory or the phonological processing model. They're similar. Here, reading is

30
00:03:23,800 --> 00:03:30,000
seen primarily as sounding out words. You sound out the words and then listen to

31
00:03:30,000 --> 00:03:35,440
the sentences being played in your head. And the problem with this is it doesn't

32
00:03:35,440 --> 00:03:39,640
account for a lot of research-based data related to eye movement,

33
00:03:39,640 --> 00:03:45,840
miscue analysis, priming studies, and brain imaging, among other things. It's a

34
00:03:45,840 --> 00:03:51,440
weak theory. The simple view of reading upon which the science of reading is

35
00:03:51,440 --> 00:03:58,160
built provides a very simplistic and inaccurate view of reading and doesn't

36
00:03:58,160 --> 00:04:05,600
fully explain important phenomena or account for relevant facts. Five facts

37
00:04:05,600 --> 00:04:12,400
in particular are left unaccounted for by the simple view of reading. And again,

38
00:04:12,400 --> 00:04:19,760
the science of reading, Clown Club, is based on the simple view of reading. Fact

39
00:04:19,760 --> 00:04:25,840
number one, unaccounted for. Proficient readers do not look at fully one-third

40
00:04:25,840 --> 00:04:32,320
of the words on the page. This indicates that readers are using more than the

41
00:04:32,320 --> 00:04:38,840
words and letters on the page to create meaning. They're using all sorts of other

42
00:04:38,840 --> 00:04:48,400
information, word recognition systems, semantic systems, syntactic systems,

43
00:04:48,400 --> 00:04:55,360
schematic systems. Two, proficient readers often insert words that are

44
00:04:55,360 --> 00:05:00,000
semantically or syntactically correct. Again, this points to the fact that

45
00:05:00,000 --> 00:05:06,120
information besides what's on the page is being used to recognize words and

46
00:05:06,120 --> 00:05:14,680
construct meaning. The third idea, the ratio of corticothalamic nerve fibers to

47
00:05:14,680 --> 00:05:20,960
the thalamocortical fibers is 10 to 1. In other words, during the act of

48
00:05:20,960 --> 00:05:24,960
reading, brain imaging research has shown that almost 10 times more

49
00:05:24,960 --> 00:05:32,080
information is flowing from the cortex down than from the page to the thalamus

50
00:05:32,080 --> 00:05:38,440
and up. This again shows that what's in the head is being used along with text

51
00:05:38,440 --> 00:05:45,520
clues to create meaning. The fourth fact, unaccounted for, the information from

52
00:05:45,520 --> 00:05:50,920
the cortex is used to direct the eyes during the act of reading. In other words,

53
00:05:50,920 --> 00:05:58,800
higher level processes drive or mediate lower level processes. And number five,

54
00:05:58,800 --> 00:06:04,640
and this is a fact misused and misstated by the science of reading clown

55
00:06:04,640 --> 00:06:12,920
club, never have so many, no one's so little about so much. They insist that

56
00:06:12,920 --> 00:06:18,720
proficient readers look at every letter, but proficient readers do not look at

57
00:06:18,720 --> 00:06:23,560
every letter. There's a variety of eye movement research that shows this, not

58
00:06:23,560 --> 00:06:27,320
one study but a whole bunch of it. When they don't, when they read, they don't

59
00:06:27,320 --> 00:06:33,480
even fixate on every word. The brain is filling in the blanks to give you a

60
00:06:33,480 --> 00:06:39,320
sense that you're seeing every word and every letter. All right, the second

61
00:06:39,320 --> 00:06:47,040
theory is called the top down theory. And I include this for a reason. Many of the

62
00:06:47,040 --> 00:06:53,240
clowns in the science of reading clown club keep misrepresenting balanced

63
00:06:53,240 --> 00:07:00,440
literacy as top down, but a lack of knowledge doesn't seem to be getting in

64
00:07:00,440 --> 00:07:06,200
the way of their coming to a conclusion about things. The top down theory

65
00:07:06,200 --> 00:07:12,120
states that during the act of reading, readers use primarily what's in their

66
00:07:12,120 --> 00:07:19,240
head to identify words and make sense of text. Information flows mostly down.

67
00:07:19,240 --> 00:07:24,160
Top down teachers would want students to have a whole lot of knowledge about

68
00:07:24,160 --> 00:07:29,880
what they're reading and memorize a whole bunch of words. And I know a very few

69
00:07:29,880 --> 00:07:37,120
if any that would adhere to this theory today. There are many who inaccurately

70
00:07:37,120 --> 00:07:42,880
call whole language and balanced literacy and psycholinguistic models of

71
00:07:42,880 --> 00:07:51,200
reading top down theories. This would be very, very wrong. Top down theories of

72
00:07:51,200 --> 00:07:58,520
reading are not in alignment with the principles of balanced literacy. So get

73
00:07:58,520 --> 00:08:04,320
rid of that one. The third theory is the interactive theory. The interactive

74
00:08:04,320 --> 00:08:09,880
theory of reading states that what is in the head interacts with what's on the

75
00:08:09,880 --> 00:08:17,480
page to create meaning. Letters, words and sentences interact with what's in the

76
00:08:17,480 --> 00:08:23,640
student's head to create meaning. Louise Rosenblatt called this a trans-actional

77
00:08:23,640 --> 00:08:30,680
theory. And in any transaction, both parties give something to get something.

78
00:08:30,680 --> 00:08:37,920
So this theory states that reading is a transaction between the reader and the

79
00:08:37,920 --> 00:08:43,960
text. And this is an alignment with schema theory and it supports the idea

80
00:08:43,960 --> 00:08:49,600
that we perceive the salient elements of reality and fill in the blanks with what's

81
00:08:49,600 --> 00:08:55,720
in our head. This is in alignment with constructivist learning theory which

82
00:08:55,720 --> 00:09:00,400
states what we use what's in our head to construct a view of things and our

83
00:09:00,400 --> 00:09:05,120
understanding of things. This is also in alignment with the neurocognitive model

84
00:09:05,120 --> 00:09:12,200
of reading or the psycholinguistic model of reading. Now these two models account

85
00:09:12,200 --> 00:09:18,560
for the unaccounted facts described earlier. These theoretical models define

86
00:09:18,560 --> 00:09:24,360
reading as the process of creating meaning with print. And during this

87
00:09:24,360 --> 00:09:31,640
meaning-making process, the brain uses interactive systems to recognize words on

88
00:09:31,640 --> 00:09:37,280
the page. Interactive and interconnected systems, phonological systems,

89
00:09:37,280 --> 00:09:44,120
semantic systems, syntactic systems, schema systems, and pattern recognition

90
00:09:44,120 --> 00:09:50,240
systems. These are all interacting and interrelated and they're used to

91
00:09:50,240 --> 00:09:55,440
recognize words. And recognizing words means that you see them and you

92
00:09:55,440 --> 00:10:01,200
automatically know what they are. And this is different from identifying words.

93
00:10:01,200 --> 00:10:05,960
And the science of reading Clown Club does not differentiate between

94
00:10:05,960 --> 00:10:12,520
recognizing words and identifying words. Identifying words is a strategy you

95
00:10:12,520 --> 00:10:19,240
consciously apply when you don't recognize a word. During the process of

96
00:10:19,240 --> 00:10:26,240
reading our eyes fixate on approximately 60% of the words on the page. And of

97
00:10:26,240 --> 00:10:33,400
these fixated words, our eyes usually stop at only one or two letters. And since

98
00:10:33,400 --> 00:10:40,960
we can perceive only those things upon which our eyes fixate, it's clear that

99
00:10:40,960 --> 00:10:48,640
our brain fills in the blanks to create meaning during the reading process. It's

100
00:10:48,640 --> 00:10:54,120
clear as well that the eyes are directed by information in the cortex much more

101
00:10:54,120 --> 00:11:01,040
so than information on the page during the process of reading. Proficient readers

102
00:11:01,040 --> 00:11:07,520
use minimal letter clues to recognize words during reading. And as stated

103
00:11:07,520 --> 00:11:13,840
earlier, our brain uses interactive and interconnected systems, phonological,

104
00:11:13,840 --> 00:11:19,640
semantic, syntactic, schema, pattern recognition systems to fill in the blanks

105
00:11:19,640 --> 00:11:24,960
as we read giving the impression that we are seeing every word and perceiving

106
00:11:24,960 --> 00:11:32,120
every letter. So, what isn't research? We talk about the science of reading and

107
00:11:32,120 --> 00:11:39,440
reading science. Oftentimes, data is collected under the guise of research.

108
00:11:39,440 --> 00:11:45,040
Data is collected and then members of the clown club will happily explain that

109
00:11:45,040 --> 00:11:52,160
research supports whatever silly claim they wish to make. But anyone can collect

110
00:11:52,160 --> 00:11:59,200
data. Collecting data is not science, it's not research, it's not reading

111
00:11:59,200 --> 00:12:07,760
science, collecting data. Science is not science unless or until it has been

112
00:12:07,760 --> 00:12:15,360
subjected to blind peer review and published in an academic journal. Now,

113
00:12:15,360 --> 00:12:20,080
when someone has conducted a research study, the process is this. They write it

114
00:12:20,080 --> 00:12:26,760
up and send it out for blind peer review. That means the reviewers don't know who

115
00:12:26,760 --> 00:12:33,080
wrote the study or did the study. Reviewers are experts in the field on

116
00:12:33,080 --> 00:12:39,320
which they are reviewing. They check for the theoretical context, the methodology,

117
00:12:39,320 --> 00:12:43,920
did they get the results right, the analysis right, and are they making

118
00:12:43,920 --> 00:12:49,720
up to conclusions based on the data. And then they vote either to accept the

119
00:12:49,720 --> 00:12:59,560
study, ask for revisions, or reject the research article. So, here is what is not

120
00:12:59,560 --> 00:13:06,960
research. And this is often used to say research supports that. Collecting data is

121
00:13:06,960 --> 00:13:12,160
not research. Anyone can collect data. The for-profits are collecting data and

122
00:13:12,160 --> 00:13:17,560
then they say for a research supports that. And research is not research unless

123
00:13:17,560 --> 00:13:23,520
and until it has been subjected to blind peer review and published in an

124
00:13:23,520 --> 00:13:30,840
academic journal. Siting a famous person is not research. Oftentimes, when I

125
00:13:30,840 --> 00:13:36,880
challenge a statement and ask to see the research, someone says, well, read so and

126
00:13:36,880 --> 00:13:40,960
so or they give me an article where a lot of things are cited. That's not

127
00:13:40,960 --> 00:13:48,880
research. If you think whole language has been debunked, send me a research

128
00:13:48,880 --> 00:13:54,440
study, a debunking study that compares whole language to something else.

129
00:13:54,440 --> 00:14:00,200
Think tanks and white papers are not research. They have not been subjected

130
00:14:00,200 --> 00:14:07,440
to blind peer review. And sadly, the fifth one, these have the most impact, but

131
00:14:07,440 --> 00:14:13,840
they are not research. American public media, Education Week, National

132
00:14:13,840 --> 00:14:20,280
Council on Quality Teachers, teaching, these are not research. Reporters,

133
00:14:20,280 --> 00:14:28,600
anyone puts this together. It's not research. And the last thing, one study

134
00:14:28,600 --> 00:14:35,120
is not the research. Often what happens under the guise of reading science is

135
00:14:35,120 --> 00:14:41,640
that one study or one article or one report is used to justify a policy

136
00:14:41,640 --> 00:14:46,640
program or practice. The clowns will flutter their white glove,

137
00:14:46,640 --> 00:14:53,200
the little fingers in the air and shout, the research says, but this isn't how

138
00:14:53,200 --> 00:15:02,000
science works. No single study should be the final word on anything. Now,

139
00:15:02,000 --> 00:15:06,600
research is like citing in your deer hunting rifle. And before you freak out,

140
00:15:06,600 --> 00:15:11,320
I grew up in a rural Wisconsin area where deer hunting was part of the culture.

141
00:15:11,320 --> 00:15:14,920
And this is the best analogy. I'm sorry I'm using it. But every fall,

142
00:15:14,920 --> 00:15:19,800
we'd cite in our rifles. And this means over the winter, the scope could get

143
00:15:19,800 --> 00:15:25,360
bumped. So, we need to adjust the scope to see if the bullet actually goes

144
00:15:25,360 --> 00:15:31,320
where the crosshairs indicate it will. Now, when you cite in your rifle,

145
00:15:31,320 --> 00:15:37,000
you don't take a single shot and say, oh, there's where my rifle is shooting.

146
00:15:37,000 --> 00:15:43,560
You take a bunch of shots to get a general sense of where it's shooting,

147
00:15:43,560 --> 00:15:49,200
a general pattern. And the more shots you take, the more confidence

148
00:15:49,200 --> 00:15:55,240
you have in your data, in your bullet shooting data. And you realize there's

149
00:15:55,240 --> 00:15:59,720
always going to be outliers. Your hand could have been shaking or you didn't

150
00:15:59,720 --> 00:16:05,040
pull the trigger just right. You can't look at a single research study and

151
00:16:05,040 --> 00:16:12,560
says, that's it. This proves it once and for all. That's a naive understanding

152
00:16:12,560 --> 00:16:17,640
of research. You need to look at a wide body of research,

153
00:16:17,640 --> 00:16:26,240
wide different types of research to get a general sense of things. One study is

154
00:16:26,240 --> 00:16:34,560
not the research. The research actually doesn't exist. Now, yes, phonics

155
00:16:34,560 --> 00:16:42,400
instruction is effective in helping children decode words. Everybody agrees

156
00:16:42,400 --> 00:16:49,720
with this. Everybody. But in these studies supporting phonics instruction, you

157
00:16:49,720 --> 00:16:55,440
have to look closely at the study to determine for what purpose, for who, how,

158
00:16:55,440 --> 00:17:02,520
when and for how long. Yes, teaching phonics is going to lead to higher scores

159
00:17:02,520 --> 00:17:08,600
on phonics measures when compared to not teaching phonics. Absolutely. Everyone

160
00:17:08,600 --> 00:17:17,440
knows that. No one disputes this. But does that phonics bump transfer to authentic

161
00:17:17,440 --> 00:17:23,160
reading experience? This does not mean that phonics is the problem for all

162
00:17:23,160 --> 00:17:28,640
things related to reading. This does not mean that phonics is the answer for all

163
00:17:28,640 --> 00:17:34,320
things related to reading. This does not mean that phonics should be taught in

164
00:17:34,320 --> 00:17:41,360
the same way to everybody all the time. But this is what the science of reading

165
00:17:41,360 --> 00:17:47,760
clown club is pushing. More phonics. Phonics is the problem and phonics

166
00:17:47,760 --> 00:17:55,760
therefore is the answer. But let's finish up with the National Reading Panel

167
00:17:55,760 --> 00:18:02,520
Report. They use the gold standards as they say. They cited when you actually

168
00:18:02,520 --> 00:18:08,880
read the report and look at their studies that phonics has a limited impact.

169
00:18:08,880 --> 00:18:13,320
This is from the report. Phonics instruction failed to produce a

170
00:18:13,320 --> 00:18:18,600
significant impact on the reading performance of low achievers in second

171
00:18:18,600 --> 00:18:25,080
through sixth grade. But what do we do? We give struggling readers more phonics.

172
00:18:25,080 --> 00:18:32,120
Title one, this stuff. The impact of phonics on comprehension is limited.

173
00:18:32,120 --> 00:18:38,320
Phonics instruction contributed only weekly, if at all, in helping poor readers

174
00:18:38,320 --> 00:18:44,520
apply these skills to reading actual text. You do phonics instruction there.

175
00:18:44,520 --> 00:18:51,200
You read over here. There's no transfer or little transfer. The Reading Panel

176
00:18:51,200 --> 00:18:55,480
Report says there were insufficient data to draw any conclusions about the effect

177
00:18:55,480 --> 00:19:00,920
of phonics instruction with normally developing readers above grade one.

178
00:19:00,920 --> 00:19:07,320
Normally developing readers above grade one. Phonics instruction fails to exert

179
00:19:07,320 --> 00:19:11,960
a statistically significant impact on poor readers in second through sixth grade.

180
00:19:11,960 --> 00:19:17,680
The majority of effect sizes for phonics instruction involved reading single words

181
00:19:17,680 --> 00:19:24,680
out of context. That's different from reading. Reading single words. Systematic

182
00:19:24,680 --> 00:19:29,080
phonics instruction should be integrated with other reading instruction to

183
00:19:29,080 --> 00:19:37,800
create a, wait for it, wait for it, balanced reading program. Unquote. Programs

184
00:19:37,800 --> 00:19:42,600
that focus too much on phonics with little time spent practicing reading are

185
00:19:42,600 --> 00:19:50,480
likely to be ineffective. This does not mean that we should not teach phonics.

186
00:19:50,480 --> 00:19:56,720
Everyone agrees that phonics instruction should be taught, should be a part of

187
00:19:56,720 --> 00:20:02,880
every emergent and beginning reading program. It's not the what of phonics

188
00:20:02,880 --> 00:20:09,920
that is in dispute. It's the how and the how much. This has been the Reading

189
00:20:09,920 --> 00:20:27,320
Instruction Show. I am your host as always, Dr. Andy Johnson.