Document Type

Thesis - University Access Only

Award Date


Degree Name

Master of Science (MS)

Department / School

Health, Physical Education, and Recreation


The purpose of the study is to establish a regression equation for predicting maximal bench press strength using the relationship with the weight lifted for one repetition (1 RM) and submaximal repetition to fatigue (REPS), the weight lifted during the submaximal test (SUBWT), body mass (BW), height (HT), bi-acromial breadth (BAB), arm length (AL), tricep skinfold (TS) and cross sectional area (CSA) of the upper arm of untrained females. In addition, a cross validation of three commonly used equations by comparing predicted 1 RM and actual I RM bench press weight was performed using this specific population. Subjects were 57 females recruited with no upper body strength training within three months prior to this study. The HT, BW, AL, UAC, TS, and BAB were measured on all subjects. The CSA was calculated from TS and UAC using an estimate according to Mayhew. Four sessions were conducted to familiarize each subject with the bench press using free weights. During these sessions, proper technique was taught using an educational video from the NSCA with instruction from the researcher. The weight lifted for a four to eight-repetition maximum ( 4-8 RM) was determined after the completion of the familiarization sessions. Five pound increments were used in the assessment. At least 24 hours following the 4-8 RM test, the maximum weight for one repetition was determined using procedure according to Kraemer. Step wise regression analysis indicated SUBWT, REPS, and BAB as the variables to be used in the prediction equation for this population. The three component prediction equation derived was: I-RM = l.223(SUBWT) + 0.727(REPS) -0.1453(BAB) - 0.1 53, (TE= ±1 .74, r2 = .90). The SUBWT and REPS from this population were then entered into regression equations from Bryzcki, Epley, and Landers. The paired t-test comparisons of actual vs. predicted indicated statistical difference using the equation by Landers (E= ±2. 11 kg, t = -3. 85, p< .001) and Brzycki (E= ±2.30 kg, t = -5.17 p< .001), but no differences using the equation by Epley (E= ±1.90 kg, t = -0.68, p = .50). This study concluded that a regression equation for untrained females provided a good estimate of bench press strength and improved the accuracy of prediction over the equation by Landers and Brzycki. In addition, the total error was lower using this equation over all three equations used in the comparison. The derived equation provides an option for exercise professionals when predicting bench press strength in untrained females.

Library of Congress Subject Headings

Muscle strength -- Measurement

Weight lifting

Weight training for women



Number of Pages



South Dakota State University