Written and maintained by CASRAI Editorial Board
Published
On this page
Definition in One Paragraph
The Test of Word Reading Efficiency (TOWRE), currently in its second edition as the TOWRE-2, is a nationally normed, individually administered psychometric instrument developed by Joseph K. Torgesen, Richard K. Wagner, and Carol A. Rashotte and published by PRO-ED to measure an individual’s ability to pronounce printed words and nonwords accurately and fluently. Normed for examinees aged 6 years 0 months through 24 years 11 months, the assessment requires 5 to 10 minutes of administration time and comprises two timed subtests: Sight Word Efficiency (SWE), which assesses the number of real printed words an examinee can recognize as whole units aloud within 45 seconds, and Phonemic Decoding Efficiency (PDE), which assesses the number of phonemically regular pseudowords an examinee can sound out accurately within 45 seconds using grapheme-to-phoneme conversion rules. Each subtest yields raw scores that convert into age-referenced standard scores (mean = 100, standard deviation = 15) and percentile ranks, which combine to generate the composite Total Word Reading Efficiency (TWRE) Index. In cognitive psychology and educational research, the TOWRE isolates word-level reading automaticity from untimed decoding mastery and general reading comprehension.
Formula With Every Symbol Explained
TOWRE scoring uses three formal transformations: calculating raw fluency rates, deriving standardized subtest scores, and computing composite index values with standard error bounds.
1. Raw Reading Efficiency Rate: Articulation throughput per unit of elapsed time:
R = C / T
- R = Reading efficiency rate in items pronounced correctly per second.
- C = Total count of words (SWE) or pseudowords (PDE) articulated accurately.
- T = Elapsed administration duration (standardized at 45.0 seconds).
2. Standardized Subtest Score (SS): Raw scores are standardized against age-band normative parameters using the z-score transformation:
z = (X – μ) / σ
SS = 100 + 15 × z
- X = Observed raw subtest score in 45.0 s.
- μ = Normative sample mean raw score for the examinee’s age band.
- σ = Normative sample standard deviation for the examinee’s age band.
- z = Standardized deviate in standard deviation units relative to peers.
- SS = Derived subtest standard score (mean = 100, standard deviation = 15).
3. Composite Total Word Reading Efficiency (TWRE) Index: The composite reading index derives from the sum of subtest standard scores:
S_sum = SS_SWE + SS_PDE
TWRE = f(S_sum)
- SS_SWE = Age-referenced standard score on Sight Word Efficiency.
- SS_PDE = Age-referenced standard score on Phonemic Decoding Efficiency.
- S_sum = Sum of bivariate standard scores.
- f(S_sum) = Normative lookup function from the TOWRE-2 manual mapping bivariate sums to a composite standard distribution (mean = 100, standard deviation = 15), accounting for inter-subtest correlation (r ≈ 0.70 to 0.80).
4. Standard Error of Measurement (SEM) and 95% Confidence Interval: Score precision is bounded using classical test theory:
SEM = SD × √(1 – r_xx)
CI_95 = SS ± 1.96 × SEM
- SD = Population standard deviation (15).
- r_xx = Score reliability coefficient (typically 0.90 to 0.94).
- SEM = Standard error of measurement quantifying random variance.
- CI_95 = Interval containing the examinee’s true score with 95% certainty.
Worked Example With Real Numbers and the Arithmetic Shown
Consider an evaluation of a 9-year-old student (chronological age 9:2) administered Form A of the TOWRE-2.
Step 1: Record Raw Results: In the 45.0-second trials, the student correctly read 62 real words on SWE (X_SWE = 62) and 34 pseudowords on PDE (X_PDE = 34).
Step 2: Calculate Throughput Rates:
- SWE Rate: R_SWE = 62 / 45.0 = 1.38 words/s (82.7 words/min).
- PDE Rate: R_PDE = 34 / 45.0 = 0.76 nonwords/s (45.3 nonwords/min).
Step 3: Convert to Age-Based Standard Scores: From normative tables for age 9:0–9:3 (μ_SWE = 60.0, σ_SWE = 10.0; μ_PDE = 32.0, σ_PDE = 8.0):
- SWE: z = (62 – 60.0) / 10.0 = +0.20 → SS_SWE = 100 + 15 × (+0.20) = 103 (58th percentile).
- PDE: z = (34 – 32.0) / 8.0 = +0.25 → SS_PDE = 100 + 15 × (+0.25) = 104 (61st percentile).
Step 4: Compute Composite TWRE Index and Confidence Intervals: Summing subtests yields S_sum = 103 + 104 = 207, mapping to a TWRE Index of 104 (61st percentile). With reliability r_xx = 0.92:
- SEM = 15 × √(1 – 0.92) = 15 × √(0.08) = 4.24
- 95% Confidence Interval: 104 ± (1.96 × 4.24) = 104 ± 8.31 = [95.7, 112.3], rounding to [96, 112].
The student’s true reading efficiency falls within the average range between 96 and 112 with 95% certainty.
Interpretation (What the Result Means, With Reading Guidance)
TOWRE interpretation involves examining absolute performance against national norms and evaluating intra-individual discrepancies between subtests:
| Standard Score | Percentile | Descriptor | Diagnostic Guidance |
|---|---|---|---|
| ≥ 131 | ≥ 98 | Very Superior | Exceptional fluency; verify potential ceiling boundaries. |
| 121–130 | 92–97 | Superior | Advanced sight-word access and rapid phonological assembly. |
| 111–120 | 76–91 | Above Average | Robust automaticity exceeding age-expected benchmarks. |
| 90–110 | 25–75 | Average | Normal, age-appropriate word reading proficiency. |
| 80–89 | 9–24 | Below Average | Mild fluency deficit; indicates need for progress tracking. |
| 70–79 | 3–8 | Poor | Moderate-to-severe deficit typical of dyslexia profiles. |
| < 70 | < 3 | Very Poor | Severe impairment in rapid lexical retrieval and decoding. |
Subtest discrepancies clarify underlying cognitive mechanisms:
- SWE lower than PDE: Points to an orthographic retrieval deficit; the examinee decodes phonetically but lacks sight-word automaticity.
- PDE lower than SWE: Classic phonological core deficit; the examinee relies on visual memorization but struggles with novel decoding.
In research design, establishing convergent and discriminant validity ensures subtests measure reading automaticity distinct from general intelligence. Furthermore, examining floor and ceiling effects confirms that timed tasks prevent adult ceiling compression while alerting researchers to potential floor compression in emergent readers.
Assumptions and Common Mistakes
Administering the TOWRE validly assumes exact 45.0-second timing with a calibrated stopwatch, adherence to standardized pronunciation rubrics, normal sensory and speech capabilities, and demographic alignment with the normative sample.
Common researcher and practitioner errors include:
- Confusing Timed Efficiency With Untimed Accuracy: Conflating speeded tasks with untimed instruments obscures fluency bottlenecks in reading acquisition.
- Comparing Longitudinal Raw Scores: Tracking unstandardized raw counts over time risks confounding developmental growth with intervention gains; researchers must account for maturation threats to internal validity in longitudinal trials.
- Scoring Inconsistency Under Time Pressure: High articulation rates demand rigorous scoring calibration; maintaining high intra-rater reliability through audio checks minimizes classification error.
- Omitting Measurement Error: Reporting point estimates without error bounds overstates precision; incorporating the standard error of the mean and SEM ensures rigorous statistical reporting.
How to Report It (APA Example)
In manuscripts adhering to APA 7th style, document the edition, subtests, alternate forms, normative metrics, and test statistics:
Method Section Example:
Reading automaticity was evaluated using the Test of Word Reading Efficiency–Second Edition (TOWRE-2; Torgesen et al., 2012) Form A. Participants completed the 45-s Sight Word Efficiency (SWE) and Phonemic Decoding Efficiency (PDE) subtests. Raw scores were converted to age-referenced standard scores (M = 100, SD = 15) and combined into the composite Total Word Reading Efficiency (TWRE) Index. Standardized alternate-form reliability coefficients exceed .85 across target age cohorts.
Results Section Example:
Participants demonstrated significant gains in overall word reading efficiency following intervention, with TWRE composite scores increasing from baseline (M = 98.40, SD = 13.85, 95% CI [94.50, 102.30]) to posttest (M = 106.20, SD = 12.10, 95% CI [102.80, 109.60]), t(49) = 4.82, p < .001, Cohen’s d = 0.60. Gains were pronounced on Phonemic Decoding Efficiency, t(49) = 5.14, p < .001, d = 0.68, reflecting enhanced sublexical phonological assembly fluency.
How to Calculate It in Excel, R or Python
The snippets below show how to compute TOWRE efficiency rates, standard scores, and confidence intervals across Excel, R, and Python.
1. Excel Formulas:
# Fluency Rate (items/sec)
=A2 / 45
# Standardized Subtest Score (M=100, SD=15)
=ROUND(100 + 15 * ((A2 - $D$2) / $E$2), 0)
# 95% Confidence Interval Half-Width
=1.96 * (15 * SQRT(1 - $F$2))
2. R Implementation:
calc_towre <- function(raw_swe, raw_pde, m_swe, sd_swe, m_pde, sd_pde, rel = 0.92) {
ss_swe <- round(100 + 15 * ((raw_swe - m_swe) / sd_swe))
ss_pde <- round(100 + 15 * ((raw_pde - m_pde) / sd_pde))
twre <- round((ss_swe + ss_pde) / 2)
sem <- 15 * sqrt(1 - rel)
ci <- c(round(twre - 1.96 * sem, 1), round(twre + 1.96 * sem, 1))
list(Rate_SWE = round(raw_swe / 45, 2), Rate_PDE = round(raw_pde / 45, 2),
SS_SWE = ss_swe, SS_PDE = ss_pde, TWRE = twre, CI_95 = ci)
}
res <- calc_towre(62, 34, 60.0, 10.0, 32.0, 8.0)
print(res)
3. Python Implementation:
import math
def calc_towre(raw_swe: int, raw_pde: int, m_swe: float, sd_swe: float,
m_pde: float, sd_pde: float, rel: float = 0.92) -> dict:
ss_swe = int(round(100 + 15 * ((raw_swe - m_swe) / sd_swe)))
ss_pde = int(round(100 + 15 * ((raw_pde - m_pde) / sd_pde)))
twre = int(round((ss_swe + ss_pde) / 2.0))
sem = 15.0 * math.sqrt(1.0 - rel)
ci95 = (round(twre - 1.96 * sem, 1), round(twre + 1.96 * sem, 1))
return {"Rate_SWE": round(raw_swe / 45.0, 2), "Rate_PDE": round(raw_pde / 45.0, 2),
"SS_SWE": ss_swe, "SS_PDE": ss_pde, "TWRE": twre, "CI_95": ci95}
metrics = calc_towre(62, 34, 60.0, 10.0, 32.0, 8.0)
print(metrics)
Frequently Asked Questions
What is the difference between TOWRE and TOWRE-2?
The TOWRE-2 (2012) added four alternate forms (A–D) instead of two, updated normative data using 2008–2009 U.S. Census demographics, and refined diagnostic sensitivity for progress monitoring in Response to Intervention frameworks.
What do the Sight Word Efficiency (SWE) and Phonemic Decoding Efficiency (PDE) subtests measure?
Sight Word Efficiency measures the speed of recognizing familiar real words as whole orthographic units. Phonemic Decoding Efficiency measures the speed of applying grapheme-to-phoneme rules to decode unfamiliar pseudowords.
How long does administering the TOWRE take?
The TOWRE-2 requires 5 to 10 minutes total. Each subtest involves brief practice items followed by a strictly timed 45.0-second testing window.
What is considered an average score on the TOWRE?
TOWRE-2 standard scores have a mean of 100 and standard deviation of 15. Scores between 90 and 110 represent the Average range (25th to 75th percentiles of age-matched peers).
Can the TOWRE diagnose dyslexia on its own?
No single test provides a standalone dyslexia diagnosis. The TOWRE-2 isolates word-level fluency and decoding speed, but clinical diagnosis requires comprehensive assessment alongside language comprehension, cognitive processing, and untimed academic measures.
Sources
- Torgesen, J. K., Wagner, R. K., & Rashotte, C. A. (2012). Test of Word Reading Efficiency–Second Edition (TOWRE-2): Examiner’s Manual. Austin, TX: PRO-ED. https://www.proedinc.com/ (accessed 2026-11-11).
- National Institutes of Health / PhenX Toolkit. (2026). Protocol 031301: Word Decoding – Test of Word Reading Efficiency (TOWRE / TOWRE-2). https://www.phenxtoolkit.org/protocols/view/31301 (accessed 2026-11-11).
- Pearson Clinical Assessment. (2026). Test of Word Reading Efficiency | Second Edition (TOWRE-2) Administration and Scoring Guide. https://www.pearsonassessments.com/store/usassessments/en/Store/Professional-Assessments/Academic-Learning/Reading/Test-of-Word-Reading-Efficiency-%7C-Second-Edition/p/100000305.html (accessed 2026-11-11).
- Torgesen, J. K., Wagner, R. K., & Rashotte, C. A. (1999). TOWRE: Test of Word Reading Efficiency. Austin, TX: PRO-ED. https://www.proedinc.com/ (accessed 2026-11-11).







