The short- and medium-term effects of an early literacy tutoring intervention

The intended frequency and duration of the program is 3-4 times a week for 8-10 weeks, over a total of 30-35 sessions where each session lasts 10-15 minutes. However, this dosage can and should be adapted to the individual child (i.e., students with greater difficulties may need more sessions). Instruction is given one-to-one (for the students with the greatest difficulties) or in small-groups of 2-3 students by a tutor. Teachers, special educators, pedagogues, and teacher’s aides may serve as tutors. The participating schools may choose the type of tutors that is best suited given their organizational context. The tutors receive a manual and can see three short films illustrating the method.

The aim of Læseklar is threefold: First, to make the students realize that words denote both content (e.g., “lion” represents the animal) and has a form made up of a number of phonemes. Second, to help students grasp the connection between letters and sounds. Third, and most important, to show students how to decode simple words (i.e., to translate written letter combinations into sounds).

The program includes three types of sessions. In the first type, the tutor uses clay figurines representing 24 Danish language sounds and letters, a box with a compartment for each letter, and a laminated sheet with the fingerspelling alphabet. A session starts with the tutor selecting three figurines. The student’s (or students’, if the session is conducted in a small group) task is to help the figurine to move into its designated compartment, or “house”. To do so, the student needs to figure out a code. The first part of the code is the sound of the first letter in the figurine’s name, for example, /l/ in “lion”, and the second part consists of fingerspelling the letter. The student should make the /l/ sound, while at exactly the same time fingerspelling the letter. Then the student moves the figurine into its house, which is marked with the corresponding letter. After the three figurines have moved into their houses, the procedure is repeated once and then the session ends. During the next session, the student repeats the previously completed figurines, and then continues with a couple of new ones. For students experiencing difficulties remembering letter-sound combinations, tutors can use a mirror so that students see how their mouths look and feel when saying letter names and sounds. This multi-sensory method can also be used in the second and third type of session.

The second type of session starts after the student has learnt about 12-15 of the letter sounds, which typically takes less than a month. The tutor uses images representing short words, which are glued onto the back of the pages in a small booklet. The tutor acts as a secretary for the student, who sounds out the phonemes in the word represented by the image. As the student sounds out the phonemes, the tutor writes the word on the front of the page. If needed, the student uses fingerspelling to aid the memory of the letter-sound combinations. Using the booklet, the student can practice by themselves or with their parents, as it is possible to verify whether words are read correctly by simply turning the page and check the image.

The third type of session aims to build decoding speed and fluency by practicing the decoding of short syllables with the help of flashcards. The student’s task is to decode as many syllables as possible during one minute. The tutor registers the result on a chart and then the same syllables are repeated in the same order two more times in the same session. This procedure usually yields a clear improvement. Making improvements visible for the students with the help of the chart usually motivates them to practice more. The second and third type of session are often used interchangeably, which decreases the risk that the student gets bored. It is also possible to go back to the first type of session, when words contain letter-sounds that were not among the first 12-15 that the students learned.

2023-01-09

2024-06-30

Our primary post-intervention outcome at the end of kindergarten will be a measure of decoding skills from a subtest of a standardized dyslexia test widely used in Denmark (the 30 real words in “Elbros ordlister” from “Ordblinderisikotesten”; Gellert & Elbro, 2016). The test score is the number of correctly read words (i.e., a number between 0 and 30).

We will use the compulsory Danish national tests in reading conducted in the first semester of second grade as our primary outcome at the first follow-up test. This test is currently mandatory for all public schools. Further compulsory national tests in reading are conducted in 3rd, 4th, 6th, and 8th grade. We will analyze these tests as well, should we have the resources.

At all measurement occasions, we will use subscales from the national survey of student well-being as secondary outcome measures. All students in public schools take this survey annually from Kindergarten to 3rd grade (Ministry of Education, 2018b).

The experimental design is a cluster-randomized controlled trial in which schools are randomly assigned to either the treatment or the control group. We prioritized inviting schools that were likely to have many at-risk students. To determine the eligible set of schools, we used data on educational characteristics of parents that are highly correlated with the national reading test in 2nd grade. Specifically, for the first round of invitations, we selected the schools in Denmark in which no more than half of the parents have a tertiary education (using data from the kindergarten classes in the school year 2019/2020). This cutoff yielded an eligible sample of 394 public schools (out of 1,135 for which we could obtain sufficient information). Our targeted number of participating schools is 80.

For administrative reasons, we will recruit over at least two school years. In spring 2022, we started inviting schools and in total 16 schools agreed to participate and were randomized to the treatment and control groups. In the second year, we will not-yet-invited schools and schools that declined the invitation in the first year. Depending on the response rate, we may also have to increase the number of eligible schools. It is also possible that the response rate is higher than expected. As our budget determines the number of schools we can give the program to, should we end up with more schools than the targeted 80 we may choose to randomly assign less than 50% of participating schools to the treatment group.

We conducted the randomization for the first round and notified the schools of their treatment status mid-November 2022. We will use a similar procedure in 2023. The treatment schools are expected to implement the program in the spring semester of 2023 or of 2024, which will give schools at least two months to prepare the implementation (and usually more, as they do not have to start early January). That is, regardless of when their school was recruited, students in treatment schools will train with the program in the second semester of kindergarten.

Among the schools that accepted the invitation to participate in 2022, we randomized the treatment (receipt of Læseklar) at the school level. We conducted a stratified randomization. As the recruitment procedure stretches over at least two school years and schools recruited the first year will receive the intervention before the recruitment in the second year begins, schools were stratified by recruitment year. Within each recruitment-year strata, we used a pair-matching procedure: We ranked schools by the proportion of students at risk of reading difficulties according to test scores of literacy and language skills, and then matched schools with adjacent ranks into pairs. The pre-test scores comes from language and literacy tests conducted by all Danish schools at the beginning of the first semester of kindergarten (typically in September or October). We then used the random number generator in Stata to determine the assignment to the treatment and control group in each pair. The assignment probability was 50% for each school. We will use the same procedure for schools recruited in 2023.

Not available

We used the random number generator in Stata.

School

Yes

80 schools

In total around 4,000 pupils: 2,000 pupils in the control schools and 2,000 pupils in the treatment schools, of which around 250 will receive training with the program.

40 treatment schools and 40 control schools.

Minimum detectable effect size is 0.36 standard deviations on the pupils receiving training with the program, and 0.20 standard deviations on peers not receiving the program.