This blog post is part of our Research Article of the Month series. For this month, we highlight “A Meta-Analysis on the Optimal Cumulative Dosage of Early Phonemic Awareness Instruction,” an article published in the journal Scientific Studies of Reading in 2024. Important words related to research are bolded, and definitions of these terms are included at the end of the article in the “Terms to Know” section.

Why Did We Pick This Paper?

Phonemic awareness (PA) is the understanding of, and the ability to manipulate, the individual speech sounds in words, which are called phonemes. PA is critical for word reading, and it is predictive of later reading and spelling ability. 

Although the basics of PA are acquired in early childhood before formal schooling, explicit PA instruction can further solidify children’s PA skills. The amount of formal PA instruction students receive may affect their PA outcomes. But how much PA instruction is enough? The researchers explore this question by examining optimal cumulative dosage. 

Cumulative dosage, measured in minutes or hours, is the product of the dose (the number of minutes per instructional session), the dose frequency (the number of instructional sessions per week), and duration (the total number of weeks of instruction). Previous studies have explored the optimal cumulative dosage of reading interventions and their impact on various reading outcomes (Roberts et al., 2022). Instead of examining reading interventions, this study aims to determine the optimal cumulative dosage of PA instruction for preschool through first-grade students. Cumulative dosage is considered optimal when the difference in effect size between the treatment group (students receiving PA instruction) and the control group (students receiving typical instruction) is the greatest. 

The optimal cumulative dosage of PA instruction may vary depending on other moderators, such as student characteristics (e.g., risk of disability status) and instructional factors (e.g., group size, whether or not letters were included in PA instruction, and the types of PA skills taught). For example, students at risk for reading disabilities may benefit from a greater cumulative dosage of PA instruction compared to students who are not at risk. 

Understanding the optimal cumulative dosage of PA instruction—and how this varies depending on student and instructional factors—can help educators allocate time and resources effectively to promote students’ reading achievement.

What Are the Research Questions or Purpose?

The researchers examined the optimal cumulative dosage of PA instruction and how participant and instructional characteristics affect the optimal cumulative dosage by addressing the following questions:

1. What is the optimal cumulative dosage of PA instruction and the maximum effect associated with that dosage on PA outcomes for preschool through first-grade students?
2. When studying group size, the presence of letters in instruction, risk for reading disabilities status, and types of PA skills as moderators, to what degree did the optimal cumulative dosage and maximum predicted effect vary from the overall optimal dosage and maximum effect predicted?

What Methodology Do the Authors Employ?

The researchers conducted a meta-analysis of 16 studies on PA instruction to estimate the optimal cumulative dosage of PA instruction required to reach the maximum effect on students’ PA outcomes.

To be included in the meta-analysis, the studies needed to:

• Use an experimental or quasi-experimental design
• Examine PA skills, using at least one measure of PA as an outcome measure
• Provide an estimate for the total cumulative dosage of PA instruction or data to compute it
• Provide pretest and posttest data to calculate effect size
• Focus on students in preschool through Grade 1
• Examine PA instruction implemented by a teacher, parent, or computer
• Be published in English

For each study, the researchers compared the change in pretest and posttest scores in the treatment group to the change in pretest and posttest scores in the control group to determine the overall effect size of PA instruction on PA outcomes. In this study, the PA outcome measure included basic PA skills such as identification (recognizing words that start with the same phonemes), blending (combining individual phonemes to pronounce a whole word), segmentation (breaking down a whole word into its individual phonemes), categorizing (sorting words based on shared or unshared phonemes), and isolation (extracting a phoneme from a word), as well as advanced PA skills such as deletion (removing a phoneme from a word) and substitution (changing a phoneme in a word to produce a different word). 

Then, the researchers calculated optimal cumulative dosage by determining when the difference in effect size between the treatment group and the control group was the greatest.

For each of the studies, the researchers noted other moderator variables that could affect the outcomes of PA instruction. These variables included:

• Group size (one-on-one, small group, or whole class)
• PA skills taught (basic PA skills or advanced PA skills)
• Use of letters (yes or no)
• Risk status (at risk or low risk)

They conducted a separate moderator analysis in order to determine the extent to which the maximum effect size and optimal cumulative dosage changed in response to each moderator variable. 

What Are the Key Findings?

Research Question 1: What is the optimal cumulative dosage of PA instruction and the maximum effect associated with that dosage on PA outcomes for preschool through first-grade students?

The maximum effect size for PA instruction was d = 0.74, and this effect was observed to occur at 10.2 hours. In other words, the estimated effect of PA instruction increases until 10.2 hours, at which point, it plateaus and decreases. As students develop automaticity in PA skills, they may reach a point after which additional PA instruction no longer yields significant PA gains. The results from this study indicate that this may occur after 10.2 hours. However, the researchers acknowledge that, due to individual differences, the optimal amount of PA instruction will differ by student, and these findings should not be taken as an oversimplified prescription for instruction. 

Research Question 2: When studying group size, the use of letters, risk for reading disabilities status, and types of PA skills as moderators, to what degree did the optimal cumulative dosage and maximum predicted effect vary from the overall optimal dosage and maximum effect predicted?

• The maximum effect of one-on-one PA instruction was larger than the overall maximum effect, but this difference was not significant. The optimal cumulative dosage of one-on-one PA instruction was 16.62 hours, which was significantly longer than the overall optimal cumulative dosage. This finding indicates that the maximum effect of PA instruction may take longer to achieve in a one-on-one setting.
• For PA instruction with letters, the optimal cumulative dosage was significantly longer (16.1 hours) than the overall model (10.2 hours). The model showed that effect size continued to increase as cumulative dosage increased, which may suggest that incorporating letters into PA instruction and PA into alphabet instruction could be an effective way to maximize PA gains. 
• The maximum effect of PA instruction for students at risk for reading disabilities was larger than the overall maximum effect, and the optimal cumulative dosage was longer. However, neither of these differences were significant. This finding was unexpected because students at risk for reading disabilities often have difficulty associating sounds with the letters or letter combinations that represent them. Therefore, they often require more intensive instruction to demonstrate PA gains. 
• For types of PA skills taught, basic or advanced, there was no significant difference in optimal cumulative dosage or effect size when compared to the overall model.

What Are the Limitations of This Paper?

It is important to note the study’s limitations when considering the applicability of its findings. First, this study focused on early PA instruction; the research sample comprised solely kindergarten through first-grade students. The conclusions derived from the analysis may not be generalizable to PA instruction for older students, particularly those in grades 2 through 5 or at the secondary level. Without further research, it is unclear whether the identified optimal dosage of PA instruction would be as effective for these older students. 

Additionally, the study did not investigate several factors that could affect the effectiveness of PA instruction on PA outcomes. Factors such as the type of instructional materials or quality of instruction could substantially influence the impact of PA instruction. The absence of these considerations in the moderator analysis suggests that the study’s advised optimal dosage may not be universally effective under varying instructional conditions. 

In summary, the research provides valuable insights into PA instruction dosage for young students in early literacy education. However, its conclusions may not be universally applicable for all grades and students, especially students at risk for reading disabilities. Further research that includes a wider range of grades, as well as moderator analysis of instructional materials and instruction quality, would be necessary to deeply understand the optimal cumulative dosage of PA instruction. 

Terms to Know

• Cumulative dosage: Cumulative dosage, measured in minutes or hours, is the product of the dose (the number of minutes per instructional session), the dose frequency (the number of instructional sessions per week), and duration (the total number of weeks of instruction). For example, if an intervention occurred for 30 minutes twice a week for eight weeks, it would have a dosage of eight hours.

• Effect size: In statistics, effect size is a measure of the strength of the relationship between two variables in statistical analyses. A commonly used interpretation is to refer to effect size as small (g = 0.2), medium (g = 0.5), and large (g = 0.8) based on the benchmarks suggested by Cohen (1988), where “g” refers to Hedge’s g, a statistical measure of effect size.

• Experimental research: Experimental research aims to determine whether a certain treatment influences a measurable outcome—for example, whether a certain instructional method influences students’ reading comprehension scores. To do this, participants are divided into two groups: an experimental group, which receives the treatment, and a control group, which does not receive the treatment. In an experimental study, these groups are randomly assigned, meaning each participant has equal probability of being in either the treatment or the control group. Both groups are tested before and after the treatment, and their results are compared.

• Meta-analysis: A meta-analysis synthesizes the results of separate studies addressing the same research question by systematically identifying and evaluating studies on a certain phenomenon, pooling their data and conducting statistical analyses, and interpreting the collective results.

• Moderators: Moderators are variables that affect the relationship between two other variables. For example, the relationship between the length of a reading intervention and reading comprehension may be stronger for students who are at risk for reading disabilities versus students who are not at risk. In this case, at-risk status would be a moderator. 

• Quasi-experimental: A quasi-experimental study is similar to an experimental study except that participants are not randomly assigned to groups. In educational research, groups often are assigned by classroom rather than through random assignment, making this kind of research quasi-experimental.

• Significant: If a study’s findings are statistically significant, it means they are unlikely to be explained by chance alone.

References

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Routledge.

Erbeli, F., Rice, M., Xu, Y., Bishop, M. E., & Goodrich, J. M. (2024). A meta-analysis on the optimal cumulative dosage of early phonemic awareness instruction. Scientific Studies of Reading, 1-26. https://doi.org/10.1080/10888438.2024.2309386

Roberts, G. J., Dumas, D. G., McNeish, D., & Coté, B. (2022). Understanding the dynamics of dosage response: A nonlinear meta-analysis of recent reading interventions. Review of Educational Research, 92(2), 209-248. https://doi.org/10.3102/00346543211051423