Journal Logo


Enhancing Muscular Qualities in Untrained Women

Linear versus Undulating Periodization


Author Information
Medicine & Science in Sports & Exercise: September 2009 - Volume 41 - Issue 9 - p 1797-1807
doi: 10.1249/MSS.0b013e3181a154f3
  • Free


Resistance training programs for improving hypertrophy, strength, and power have normally followed the concept of periodization. There are two basic periodization models, although there is disagreement over the terms given to the different models. The first is termed traditional, stepwise, or linear periodization (LP), whereas the other is termed nonlinear, nontraditional, or undulating periodization (UP) (2,24). This second model has also been termed mixed-methods resistance training (17). LP is characterized by training that starts with high-volume and low-intensity exercises, with volume reduced and intensity increased as the athlete/trainee works toward a peak, typically over a 10- to 12-wk cycle. Theoretically, the early high-volume phase emphasizes the hypertrophic adaptations (15), and the later, high-intensity phase stresses the neural responses. In contrast, UP is characterized by a type of periodization in which training volume and intensity undulate on a daily or weekly basis. Part of the rationale for UP is that prolonged LP might lead to neural fatigue, which in turn would compromise strength gains (22).

Consistent with the above hypothesis, one study has reported significantly greater gains in strength when training using UP was compared with training using LP. However, as this study failed to equate training volume between comparative programs, it is difficult to know if the observed differences were due to differences in training volume or differences in the type of periodization used. On the basis of this, other researchers have proposed that if volume (repetitions) and intensity remain equal during a training period, participants would achieve similar results regardless of the training structure used (2,25). In fact, one study has reported no differences in strength and power gains when training with LP or UP programs when training volume (repetitions) and relative intensity was equated for the entire training period (2). These results emphasize the importance of matching training volume when attempting to compare different types of periodization.

It is difficult to generalize the findings of many studies to the general population as few studies have examined the effects of periodized strength training on women. In the studies that have used female participants, similar problems regarding the matching of training volume were found. In a study (7) with female participants that matched workloads (weight lifted × repetitions × sets), no significant differences were observed between different resistance training programs. In contrast, studies that compared UP protocols with single-set circuit resistance training in untrained women found that the periodized groups improved strength and power performances earlier, and achieved continued gains in strength (10,12). However, as the UP programs in these studies had higher training volumes, the efficacy of the structure of periodization could not be ascertained.

Although most research suggests that periodization should be used during resistance training, it is not clear whether it is the varying structure of the periodized programs or the differences in training volume that makes periodization successful. Clearly, more studies on the efficacy of LP and UP are needed, especially those that assess strength changes in women. Such studies should be careful to equate training volume between training protocols because this would make it easier to assess if the differences between training programs were due to the structure of the periodized programs or because of differences in volume and intensity. Training volume, however, should include both repetitions and mass lifted because this would constitute a better estimate of work accomplished (27). It is also important to avoid confounding factors such as concurrent endurance training (11). Therefore, the purpose of this study was to examine the efficacy of two periodized programs (one that varied training intensity and volume load every 3 wk, and the other daily) in producing changes to upper and lower body strength qualities. It was hypothesized that there would be no difference between LP and UP in enhancing strength qualities in untrained females.


Experimental overview.

This study compared strength changes in response to two periodized resistance training programs (performed immediately after a 3-wk pretraining conditioning period). The LP program had 3-wk phases that emphasized hypertrophy, maximal strength, and power, respectively, whereas the UP program emphasized a different strength quality on each training day-hypertrophy on Monday, maximal strength on Wednesday, and power on Friday. Training intensities of 75%-80% of one-repetition maximum (1RM) were used for hypertrophy training, 85%-90% of 1RM for maximal strength training, and 30%-40% of 1RM for power training. Maintaining the designated intensities for each strength quality ensured that total training intensity and volume load (repetitions × mass lifted) remained similar for both groups and allowed the examination of the effects of periodization type on subsequent adaptations.


Twenty female students (mean ± SD: age 20 ± 2 yr, mass 66.1 ± 9.5 kg, height 170.7 ± 8.5 cm) who were recreationally involved in social- and amateur-level sports (basketball, soccer, netball, and hockey), but who were not systematically training, volunteered to participate in the study. None had participated in strength training in the previous 6 months, and all were asked to maintain their normal dietary and activity habits throughout the experimental period. All of the participants were nonsmokers, were not taking any medication, and had no known medical conditions or physical injuries that could confound the results of this study. Participants were informed of the potential risks and benefits associated with the investigation and gave informed consent. Approval for the study's procedures was granted by the Institutional Research Ethics Committee.

Testing procedures.

Before the first testing session, participants attended four familiarization sessions to determine the hand and feet positions and the depth that the barbell had to reach to improve reliability of position and joint angles. After the familiarization, tests were performed every 3 wk, with the first test session performed before pretraining conditioning (Fig. 1). A 48-h rest interval was given between the last training session and subsequent testing sessions to enable participants to recover from previous training. The familiarization sessions and pretraining conditioning sessions (a minimum of 12) offered the participants adequate practice time to improve test and training exercise techniques before the experimental period began (training phases I, II, and III). Changes throughout training were assessed using several tests performed in the following order to minimize fatigue: body mass, girth measurements, ultrasound imaging, 1RM squat (1RMSQ), average power during the countermovement jump (SQJpwr), 1RM bench press (1RMBP), and average power during the bench press throw (BPTpwr).

Testing and training schedule during a 12-wk period incorporating a pretraining conditioning period and three specific training phases, preceded by familiarization sessions.

Body mass and limb girth.

Body mass (kg) and height (cm) were determined using an electronic weighing scale and a wall-mounted stadiometer. Measurements of girth were taken at three sites on the right limbs; the mid-upper arm (midway between acromiale and radiale landmarks with arm hanging freely at the side, and palm facing thigh), the mid-thigh (midway between the anterior superior iliac spine and the proximal border of the patella), and the lower one-third portion of the thigh (lower one-third mark between the anterior superior iliac spine and the proximal border of the patella). The third site was made to duplicate the site used to obtain muscle CSA of the right rectus femoris.

Muscle CSA.

A B-mode ultrasound (Toshiba SSA-250A, Tokyo, Japan) was used to assess changes in the size of the right rectus femoris after training. A pilot study (n = 22) was performed before the commencement of the study to examine the operator's reliability, and an ICC value of 0.992 (P < 0.05) was obtained for intratester reliability. CSA images were collected only from the right rectus femoris as it has been suggested that, of the group of thigh muscles, only the rectus femoris exhibited a significant CSA increase after a 6-month total-body conditioning program (21). The CSA images were obtained from a site one-third of the distance from the proximal border of the patella to the anterior superior iliac spine, with the knee propped on an angled Styrofoam block to allow for knee flexion of 20°. This was performed to control the effects of joint position and state of muscle length on muscle thickness and to maintain consistency of position for all subjects. The transducer head was held perpendicular to the site without any depression occurring on the skin surface. An even and light pressure was maintained to prevent deformation of superficial structures and to limit the squeeze of gel from under the transducer head. Once the image was optimized on the monitor, it was recorded for 10 s using a video recording device: two consecutive images were recorded from the same site. The recorded images were then transferred to a computer and stored as high-resolution JPEG files. Each image was assessed for muscle CSA using the ImageJ (version 1.32j) image processing program (National Institutes of Health, Bethesda, MD), which calculated area and pixel value statistics of the defined sections on the CSA images. Each muscle image was identified and traced along the inner edge of the fibrous sheath surrounding the muscle using software tools, and the CSA was then calculated using the area calculation menu (cm2). The two measurements from each site were then averaged and used for further analysis.

Assessment of maximal strength.

Assessment of dynamic isoinertial maximal strength was performed using the 1RMSQ and 1RMBP on a modified Plyometric Power System (PPS; Plyopower Technologies, Lismore, Australia) according to previously described procedures (16). During the 1RMBP, the barbell was lowered with shoulders abducted and elbows flexed at 90°, respectively, with the wrists/hands vertically in line with the elbows until the bar was approximately 2 cm above the chest. The barbell was immediately lifted back to starting position without any bouncing action. The 1RMSQ was performed with feet shoulder width apart, and the barbell descending until the knee angle was approximately 110° before being lifted back to standing position (19). For both the 1RMBP and 1RMSQ, participants performed a warm-up set of 10 repetitions with a resistance approximating 30% of their estimated 1RM. This was followed by warm-up sets of 10 repetitions of 50% of 1RM and 5 repetitions of 75% of 1RM. A resistance approximating 3RM was then loaded on to the barbell, and the subject was asked to lift it not more than three times. On the basis of the number of successful lifts and the form shown as she performed the lifts, a 1RM load was estimated. After this, trials for 1RM were performed. Every successful trial was followed by 3-5 min of rest, with heavier loads being attempted until the 1RM was determined. The process of determining the 1RM generally took no more than four trials.

Assessment of power.

Average mechanical power was assessed during the bench press throw (BPT) and a squat jump with countermovement (SQJ) using the same PPS. The PPS is similar to those previously described (16) but differed in that force data were collected through two pretensioned piezoelectric transducers (Type 9251A; Kistler, Winterthur, Switzerland) placed in-between the original barbell and a shorter bar. As a force was produced against the barbell (during exercise performance), these transducers decomposed the force acting in any direction into three components orthogonal to one another. The signals from each of the three components were then fed into an eight-channel summing amplifier (Type Z11449; Kistler) and output as voltages. Force and displacement data were acquired at 1000 Hz during a 5-s epoch and filtered using a second-order low-pass recursive Butterworth filter with a cutoff frequency of 5 Hz and displayed on screen. When data had been collected and saved to disk, a custom-written data analysis software, Plyopower Analysis, developed using Labview 5.1 (National Instruments, Austin, TX) was used to analyze the displacement and force data. The software calculated the discrete differentiation of the filtered displacement data to compute instantaneous velocity using initial and final condition settings to minimize boundary errors. Data values of instantaneous velocities were then multiplied by force data to obtain values for instantaneous power. Although the piezoelectric transducers allowed force data to be collected in three-dimensional planes, only vertical forces (z direction) together with vertical displacement were evaluated for this study. The data analysis process took into consideration the division of movement into concentric and eccentric phases. The eccentric phase commenced from the beginning of the movement (defined as onset of bar descent) until the time before the start of the concentric phase, whereas the concentric phase was defined as the time point when bar velocity changed from negative to positive. Peak power output was taken as the highest instantaneous power produced during the propulsive phase. Power output was also averaged over the concentric phase to derive average power output.

Power assessment during the BPT and SQJ used absolute loads of 13 kg for the upper body and 22 kg for the lower body. These absolute loads were chosen on the basis of a pilot test to approximate the 30% load for optimal power in novice participants before the commencement of training. Both the BPT and SQJ involved movements that were similar to those performed during the assessment of strength, with the exception that the barbell was released at the end of the movement in a ballistic manner. No pause was permitted between the eccentric and concentric phases, and the correct depth of the barbell was visually monitored by the tester. At maximum height, the electromagnetic braking system of the PPS automatically engaged to halt the bar. Each participant performed three consecutive trials for the load tested. A rest period of 1 min was given between each explosive trial, whereas 5 min was given between different loads. The best of the three trials was analyzed for power and displacement of the barbell.

Resistance training program.

After pretraining conditioning, participants were ranked and then assigned to either the LP or the UP group on the basis of their squat index (1RMSQ / mass) ensuring that the average for each group was not significantly different at baseline. An A-B-B-A procedure was used placing the participant with the highest SQ index was placed into group A, the second and third ranked participants into group B, the fourth into group A, and so on, until all the participants had been assigned. This alternation of participants according to their rank in a performance variable helped to ensure that for each pair of participants, one group did not always get the higher score, and both groups started with almost equal means at the beginning of training. Groups A and B were then randomly put into the LP or the UP training program for 9 wk, with 10 participants in each group. Both groups trained three times per week on a Monday-Wednesday-Friday schedule. Although the BP ranking was not directly used during the participant assignment process, the BP index did not differ between groups.

Each training session took no more than 1 h, inclusive of warm-up and cool-down. Participants performed a standardized warm-up that began with riding a stationary cycle ergometer (Bicycle Ergometer; Monark, Varberg, Sweden) for 3 min using a light resistance setting of 60 W (60 rpm × 1 kpm). Immediately after this, each subject proceeded to activity-specific warm-up activities-one set of 10 repetitions using approximately half the training load. This was followed by the actual training set. At the end of every session, each subject cooled down by performing a 5-min ride on the same cycle ergometer followed by a standard full-body stretching routine. Each stretching position was held for 10 s, and two repetitions were performed for each exercise. All sessions were supervised by the researcher, and the same instructions were given to each subject. A 10-exercise regimen consisting of a mixture of core and assistance exercises using both free weights and exercise machines was used for the entire training duration. Four upper body and four lower body exercises were performed bilaterally, with an abdominal and a back exercise added to give the participants a balanced whole-body workout that would better simulate real-life training conditions (Table 1). Components of the program contained both concentric and eccentric muscle actions. Training using multiple exercises within a single training session has been postulated to induce better hormonal anabolic environment (26). Full training compliance (100%) was observed by 18 of the subjects (total of 36 training sessions). Two other subjects each missed one training session because of unforeseen circumstances and recorded an acceptable attendance rate of 97%.

Exercises, sequence, rest, and pace of movement used during training.

The bench press and the back squat were the main focus of the exercise program. The loads used for these two exercises during training were determined from the 1RM scores during testing. Subjects would train for bench press and squat at 30%-40%, 75%-80%, or 85%-90% of 1RM according to the intensity that was set in the program (Table 2). The loads remained the same until the next testing session, after which they were adjusted according to the new 1RM score. For all other exercises, subjects trained using 6RM or 10RM loads according to the intensity set for that session. If a 6RM intensity was required, subjects identified a load that allowed them to complete six repetitions with their maximal effort. All loads used were recorded so that subjects had a guide line as to which loads should be used during training. During familiarization and through the pretest conditioning period, subjects were trained to determine their 6RM or 10RM loads.

Alternation of volume and intensity for LP and UP programs.

Statistical analysis.

Before the commencement of training at T1, the two groups were statistically compared using independent t-tests for demographics and strength to examine whether the subjects differed in any significant way before training. Independent t-tests were performed on the training volume load for each week and phase of training to examine differences between programs as training progressed. Independent t-tests were also performed on the total training volume load to confirm that both groups were not significantly different for the total amount of weight lifted for the eight upper and lower body exercises (abdominal and back exercises were excluded from the analysis). After the training period, the results of each measure (13 measures in total) were analyzed by a 2 × 4 (2 groups × 4 times) mixed-factor, repeated-measures, ANOVA to compare within-groups (pretraining, between phases, and posttraining), between-groups (LP and UP), and interaction effects. If a significant interaction effect was found (P ≤ 0.05), then independent t-tests were used to locate significant between-group differences, whereas main effects for time for each group were analyzed using one-way repeated measures with Bonferroni adjustments. All statistical analyses were performed with a statistical software package (SPSS version 12.0.1; SPSS, Inc., Chicago, IL). Means, SD, confidence intervals (CI), and percent change were calculated for all test variables. The level of significance was set at P ≤ 0.05. Effect sizes (ES) were also calculated with ES of 0.2, 0.5, and 0.8 representing small, moderate, and large differences, respectively.


Pretraining demographic and strength comparisons.

No significant prestudy differences between the two training groups were detected for age, mass, height, or upper and lower body strength. Means ± SD for each of the parameters mentioned are shown in Table 3.

Pretraining demographic and strength data by group.

Training volume and intensity.

Both LP and UP groups were programmed to have the same number of training sessions (n = 27), total number of sets (n = 660), and total number of repetitions (n = 5268). Training intensity was varied on a daily basis for UP, and changed every 3 wk for LP, but the mean intensity was the same at the end of training. Training volume load was compared weekly, after each 3-wk phase, and also at the end of training for the BP and SQ combined, and for all exercises. Similar results were obtained whether the analysis was performed on all-exercise volume load or combined BP-SQ volume load. Thus, results for 1RMBP and 1RMSQ combined will not be reported. There was a significant interaction for training volume with significantly greater volume load performed by the LP group for weeks 1, 2, 3, and 6, whereas the UP group had significantly greater training volume load for weeks 7, 8, and 9 (Fig. 2). When volume load was measured by training phases, the LP group performed a significantly higher volume load during phase I, whereas the UP group had a higher volume load during phase III. No difference was observed between the two groups during phase II. Total training volume load was not significantly different for the two groups (LP: 175.6 ± 28.9 × 103 kg; UP: 157.5 ± 35.9 × 103 kg).

Training volume load (kg) by week and phase. Results represent mean ± SD. *Significantly different volume from other group (P ≤ 0.05).

Body mass and limb girth.

Body mass remained unaltered during the experimental period for both groups (Table 4). There were no significant within- or between-group main effects and no interaction effect between training program and time for changes in body mass. Mean girth scores (Table 4) also showed neither a significant main effect for group nor an interaction between time and group for both arm and thigh girth measurements. There was, however, a significant main effect of time for both girth scores. Mean arm girth at T1 was not significantly different from that of T2, but was significantly lower than T3 and T4. Significant thigh girth increments were observed at T2, T3, and T4 compared with T1.

Changes in muscle mass and limb girths during the training period.

Muscle CSA.

A significant group-by-time interaction occurred for CSA changes of the right rectus femoris during the experimental period, but there was no between-group main effect. One-way repeated-measures ANOVA found that the LP group improved muscle CSA means significantly at T2, T3, and T4 when compared with T1. There were no significant differences between muscle CSA scores at T2, T3, and T4. The same analysis performed on the UP group demonstrated that muscle CSA improved significantly from T1 to T2, T3, and T4. The T2 score was also significantly different from T3 and T4 scores, but differences between T3 and T4 scores only approached significance (P = 0.093). The LP group had the largest increase at T2 (after hypertrophy training), followed by a smaller increase at T3 (after maximal strength training), before decreasing at T4 (after power training). The UP group meanwhile produced the largest improvement at T3, with similar but smaller increments at T2 and T4. Independent t-tests found a significant difference between LP and UP only at T2. Data comparing the groups across time are listed in Table 5.

Changes in CSA of the right rectus femoris across time for the LP and UP groups.

Maximal strength.

Table 6 shows the mean, SD, and changes in percentage increases of both upper and lower body strength as measured by absolute values of 1RMSQ and 1RMBP. There was no difference in results when relative strength (strength index) was analyzed, and therefore, only absolute values are discussed. Neither significant group-by-time interactions nor significant between-group main effects could be detected for both the 1RMBP and 1RMSQ. There was, however, a main effect of time for the 1RMBP, and mean scores at T2 (mean = 42.2 kg, CI = 37.0-43.1 kg), T3 (mean = 45.2 kg, CI = 40.0-50.3 kg), and T4 (mean = 47.9 kg, CI = 42.0-53.7 kg) were significantly greater than the means at T1 (mean = 38.4 kg, CI = 33.6-43.1 kg). Similarly, there was a significant main effect of time for the 1RMSQ, with mean scores at T2 (mean = 110.5 kg, CI = 101.0-120.1 kg), T3 (mean = 124.8 kg, CI = 115.2-134.4 kg), and T4 (mean = 133.8 kg, CI = 123.7-143.1 kg) significantly greater than the means at T1 (mean = 97.0 kg, CI = 88.2-105.8 kg). Mean 1RM scores were also significantly higher at each subsequent test occasion (P = 0.0005) between all test occasions (Fig. 3).

1RM upper and lower body strength values at each time for the LP and UP groups.
Changes in 1RM bench press (BP) and squat (SQ) means between test occasions for LP and UP groups and for pooled data from both groups. Error bars, ± SD. aSignificantly greater than T1 mean; bsignificantly greater than T2 mean; csignificantly higher than T3 mean (P ≤ 0.05).

Average mechanical power.

Average mechanical power output (SQJpwr and BPTpwr) was examined during the SQJ and BPT (Table 7). The interaction between group and time was not significant. The main effect between groups was also not significant, but there was a significant main effect of time with average power increasing significantly from T1 to T2, T1 to T3, and T1 to T4. Mean BPTpwr at T1 was 208.6 W (CI = 184.9-232.3 W), at T2 was 220.8 W (CI = 193.1-248.4 W), at T3 was 228.1 W (CI = 202.6-253.6 W), and at T4 was 234.5 W (CI = 201.2-267.7 W). The mean pooled SQJpwr was 1011.3 W (CI = 922.9-1099.8 W) at T1, 1054.2 W (CI = 964.7-1143.6 W) at T2, 1076.1 W (CI = 992.4-1159.8 W) at T3, and 1111.9 W (CI = 1029.1-1194.7 W) at T4.

Upper and lower body average power output values at each time using 13 kg (BPT) and 22 kg (SQJ) for LP and UP.

Barbell height during BPT and SQJ.

No significant group × time interaction effect was detected for changes in jump and throw heights (Table 8), and there was no significant main effect for group. Both height of throw and jump recorded significant main effects for time. Throw height scores at T1 were significantly lower than each subsequent test occasion. T2 throw height scores were significantly lower than T3 and T4 throw scores. Mean throw height was 0.39 m (CI = 0.33-0.45 m) at T1, 0.46 m (CI = 0.38-0.53 m) at T2, 0.51 m (CI = 0.43-0.59 m) at T3, and 0.58 m (CI = 0.48-0.68 m) at T4. Similar results were observed for jump height with significant differences observed between T1 and all subsequent test occasions, between T2 and T4, between T3 and T4, but not between T2 and T3. Mean jump height was 0.60 m (CI = 0.53-0.66 m) at T1, 0.65 m (CI = 0.58-0.73 m) at T2, 0.64 m (CI = 0.59-0.76 m) at T3, and 0.74 m (CI = 0.65-0.84 m) at T4. Large percentage increases were observed in both groups for height of throw (LP: 56.4%; UP: 44.8%), with smaller increases observed for height of jump (LP: 28.0%; UP: 21.5%).

Jump and throw height of the barbell at each time using 13 kg (BPT) and 22 kg (SQJ) for LP and UP.


This study is the first, to our knowledge, to examine differences between LP and UP for improving strength qualities in untrained women. An important feature of the present study is that training volume load (estimated by multiplying the total number of repetitions by the mass lifted), intensity, sets, and repetitions were matched between experimental groups by the end of training. The main finding was that both LP and UP training produced significant improvements in most of the variables tested, without significant interactions between the training groups. These results suggest that the total volume load of training is more important than the variation of training volume load and intensity within a periodized program for improving strength qualities.

As previously reported in women after resistance training (8,10), muscle hypertrophy (as assessed via girth measurements and muscle CSA changes) was observed in this study. Both training programs, however, were found to be equally effective in stimulating muscle hypertrophy, with hypertrophic responses occurring earlier in the thighs (after 3 wk of conditioning and 3 wk of training) than in the arms (after 3 and 6 wk of conditioning and training, respectively). Caution is required when interpreting changes in muscle size by girth measurements because of the large technical error of measurement for this measure. Because both training programs in the present study were matched for total training volume load, this may explain why there were similar hypertrophic responses despite the variations in the way in which volume load and intensity were manipulated.

Ultrasonic imaging seems to be more sensitive to muscle CSA changes than girth measurements and is a more reliable measure of muscle hypertrophy (3). Perhaps because of this improved measurement sensitivity, a statistical interaction between LP and UP training was observed. The LP group recorded their largest increase in CSA after hypertrophy training, followed by a smaller increase after maximal strength training and a slight decrement after power training, whereas the UP group achieved hypertrophic gains after every phase (Table 5). Interestingly, these different patterns seem to reflect the variations in training volume of the two different programs. However, there was no end difference in performing all hypertrophic sessions within a 3-wk phase (LP) or doing only one hypertrophy session a week for nine consecutive weeks (UP).

Taking into consideration all subjects, the change in muscle CSA of the rectus femoris in the current study was approximately 6.8% after 3 wk of training, 11.3% after 6 wk, and 11.8% after 9 wk. These changes are larger than those reported in previous longer-duration studies (20-24 wk) that have used female subjects with similar physical attributes and training history (4,9). It has been suggested that training programs using complex (multijoint) exercises, such as the bench press and leg press, result in minimal muscle hypertrophy in the first few weeks of training because a neural adaptation period of more than 10 wk is required (4). These contrasting results may be partially attributed to the structure of the respective training programs. One study (4) may have reported smaller CSA changes because of a training frequency of only twice a week, which may be insufficient for hypertrophy, whereas another study (9) also included aerobic training, which may have impaired CSA changes (11). The subjects in the current study may also have achieved a faster rate of muscle CSA improvement during the experimental period because of the prestudy training that allowed for neural improvement to occur and stabilize, thus allowing hypertrophic improvements to begin more quickly. This prestudy conditioning was not performed in both previously mentioned studies (4,9).

Both the LP and UP training protocols recorded significant increases in maximal strength for the upper and lower body, with no statistical difference between the programs. When the current subjects were examined for percentage improvements in strength, the UP group obtained gains of 28.2% for the BP and 41.2% for the SQ, compared with 21.2% for the BP and 34.8% for the SQ by the LP group. It has been reported (24) that 12 wk of UP training produces greater percentage gains in both the bench press (LP: 14.4%, UP: 28.8%) and leg press (LP: 25.7%, UP: 55.8%) compared with LP training for male subjects, but it was not stated if there was a significant interaction effect between the two protocols. It is also interesting to note that although this study (24) reported their subjects as previously strength-trained (resistance training experience of approximately 5.2 yr), percentage increments of up to 55.8% for the leg press and up to 28.8% for the bench press were observed. These percentages are normally associated with subjects with lesser training experience (23), such as those from the current study. The same study (24) also reported that UP elicited greater strength gains in the first 6 wk of training, whereas strength gains in the second half of the 12-wk program were not significantly different from those attained by the LP program. These results contrast with those of the current study because the UP group achieved consistent gains as the study progressed, whereas the LP group achieved the largest gains after phase I or II, and decreasing gains subsequently. This trend mirrored the training volume load performed by both training groups, reinforcing the opinion that larger doses of training volume are more important for hypertrophic and strength gains than the manipulation of volume and intensity (2,25). Unfortunately, it was not possible to compare the volume load of the current study with that mentioned earlier (24) because the volume in their study was represented by repetitions.

Upper body (bench press) percentage gains in the present study are comparable to those reported in studies that have also used untrained female subjects but for a longer training period (7,9). Comparative data for the SQ were more difficult to obtain because most studies with female subjects used the leg press for lower body strength assessment, and the two studies using the SQ (7,9) had subjects perform the parallel SQ (with upper thigh parallel to floor), whereas the subjects in the current study performed the SQ to a knee angle of 110°. A reduced-range SQ (such as those performed to 110°) will produce 1RM scores that are greater than scores produced during the parallel SQ (1). Nonetheless, percentage improvements in SQ strength were found to be similar regardless of differences in knee angles. A possible explanation as to why the current group of subjects managed to obtain similar gains in strength within a shorter time frame was that the previous studies (7,9) both ran concurrent aerobic programs that may have hindered strength gains (11) because of the impaired muscle tension development brought about by fatigue. Strength improvements in the lower body were greater than those in the upper body regardless of the program used. Previous researchers have also observed this phenomenon (7,9,24,25). It was proposed that these differences between upper body and lower body strength occurred because of the smaller muscle mass involved during upper body training, which would produce smaller gains, especially during a short period (29).

An unexpected development in this study was that the LP group continued to increase strength throughout the phase III training (4.4% for the BP and 5.9% for the SQ), although the training intensity and volume load were at its lowest (30%-40% of 1RM, volume = 43,821 kg). Through the same phase, the UP group performed training with higher loads and volume load (ranging from 40% to 90% of 1RM, volume load = 60,781 kg) and achieved strength increases of 10.1% and 12.8% for the BP and SQ, respectively. Previous research with untrained males (14) has similarly observed strength developments after a low-load and low-volume training. It was proposed that (14) light loads performed with sufficiently high acceleration could produce high tension within muscle and also recruit high threshold units, thus aiding muscle strength improvement. This suggests that as with untrained men, performing rapid movements with low loads and volumes can result in strength gains for untrained women. This prescription could be suitable for novice individuals who may be anxious about training with heavier loads, especially when using free weights. Caution must be exercised, however, if prescribing light-load training because prolonged use of light loads has been suggested to be detrimental to strength and power performances (2). Furthermore, percentage gains in muscle mass and strength during phase III were greater in the UP group who also trained with a greater volume load. In spite of this, anecdotal feedback from LP subjects suggested that the light-load power training during phase III acted as a psychological relief after 6 wk of heavy training, emphasizing one of the main tenets behind the concept of periodization-alternating heavy and light training for the alleviation of physical and mental stress. UP subjects similarly felt that power training once a week provided some psychological and physical relief. The psychological benefit from reduced training stress may be worth the tradeoff of a slight decrease in muscle mass.

Whereas several periodization studies (8,24,28) have reported changes in maximal strength, little is known about the influence of periodization on power development (especially in females). In the current study, average mechanical power was assessed during BPT and SQJ activities. Although average mechanical power production improved at the end of the training period, as observed for changes in maximal strength, there was no significant difference in the efficacy of either training group. The similarities in the trends for improvements in strength and power are consistent with the suggestion that increments in power may be linked to increments in strength (14). Comparative average mechanical power scores during BPT and SQJ from other studies were not found as few studies using female subjects have assessed this component. Of those that did (9), values for peak power with relative loads were obtained. Although some studies reported average mechanical power output (1), they recruited power-trained rugby-league players, and as such, their scores were much higher than those observed in the current study. The comparison of power output with previous studies (13) is also made complicated by differences in the measurement and calculation of power, the experimental protocol, equipment, and different body positions adopted during testing.

Average power increased significantly from pretest to posttest in spite of the subjects performing traditional explosive power training without projecting (releasing from contact) the barbell or weight implement at the end of movement. Although it has been suggested that holding onto the barbell at the end of an explosive bench press action will decrease power (20), the difference in average power output between projected and nonprojected movements has been approximated to be only 5.8% (5). This difference may not be large enough to discard the use of traditional, nonprojected movements during power training, especially because equipment such as the PPS is not readily available to most individuals performing resistance training. Thus, although significant improvements in power were seen in the present study, further research is required to investigate if greater gains could be obtained by projecting the weights throughout a training period.

Although the LP group performed light-load power training exclusively during the final 3 wk of training, gains in average power production were similar to that achieved during the heavy maximal strength training performed earlier. The intention to perform all power exercises as forcefully and as explosively as possible may have provided a suitable stimulus for an increase in velocity (5) that, in turn, improved power production. It has been suggested (13) that light loads lifted explosively could provide sufficient neuromuscular facilitation to improve contractile efficiency allowing greater force production and power production. Fast contraction velocities have also been reported as the most effective for increasing muscular power (18). Power improvement, as measured by average mechanical power output, may also be influenced by the training volume load as the LP group achieved their highest power increase during the first 3 wk when volume load was the highest, and power increases tapered off with decreasing training volume load.

Maximal barbell height achieved during BPT and SQJ was used to assess the effects of the periodized programs on the functional aspects of jumping and throwing. Absolute loads of 13 kg for the BPT and 22 kg for the SQJ were associated with increasing barbell heights as the subjects became more powerful. An interesting observation is that there was an increase in the projected barbell height, although training for power was performed without projection of the resistance training implement. This form of power training may have improved the ability of the subjects to accelerate the barbell, thus improving power production and projected barbell height. As this is the first study to report changes in a loaded barbell height after training, further studies are required.


For women who participate in recreational and amateur-level sports, but have not undertaken resistance training, both LP and UP were equally adept in improving strength qualities by the end of the 9-wk training period. There were, however, differences in the relative timing of the various adaptations to training, and short-term training adaptations in response to the two periodization models were different. Improvements in hypertrophy were larger and occurred earlier than previously reported. Most of the improvement in strength and power seem to be associated with improved hypertrophic changes in muscle brought about by increased training volumes. Nonprojected, light-load, explosive training was found to be capable of bringing about small increases in strength and power. Some caution needs to be used in extrapolating the results of this study to athletes or physically stronger and more powerful female populations because most studies using untrained individuals have shown great improvements regardless of the type of training program (6). Therefore, further research using stronger women with resistance training experience is needed to extend the findings of the current study.

This study has not received funding for research from any organization or institution. The results of the present study do not constitute endorsement by ACSM.


1. Baker D, Nance S, Moore M. The load that maximizes the average mechanical power output during jump squats in power-trained athletes. J Strength Cond Res. 2001;15:92-7.
2. Baker D, Wilson G, Carlyon R. Periodization: the effect on strength of manipulating volume and intensity. J Strength Cond Res. 1994;8:235-42.
3. Bemben MG. Use of diagnostic ultrasound for assessing muscle size. J Strength Cond Res. 2002;16(1):103-8.
4. Chilibeck PD, Calder AW, Sale DG, Webber CE. A comparison of strength and muscle mass increases during resistance training in young women. Eur J Appl Physiol Occup Physiol. 1998;77:170-5.
5. Cronin J, McNair PJ, Marshal RN. Developing explosive power: a comparison of technique and training. J Sci Med Sport. 2001;4:59-70.
6. Häkkinen K, Komi PV. Changes in electrical and mechanical behavior of leg extensor muscles during heavy resistance strength training. Scand J Sports Sci. 1985;7:55-64.
7. Herrick AB, Stone WJ. The effects of periodization versus progressive resistance exercise on upper and lower body strength in women. J Strength Cond Res. 1996;10:72-6.
8. Kraemer WJ, Häkkinen K, Triplett-Mcbride NT, et al. Physiological changes with periodized resistance training in women tennis players. Med Sci Sports Exerc. 2003;35(1):157-68.
9. Kraemer WJ, Nindl BC, Ratamess NA, et al. Changes in muscle hypertrophy in women with periodized resistance training. Med Sci Sports Exerc. 2004;36(4):697-708.
10. Kraemer WJ, Ratamess N, Fry AC, et al. Influence of resistance training volume and periodization on physiological and performance adaptations in collegiate women tennis players. Am J Sports Med. 2000;28:626-33.
11. Leveritt M, Abernethy PJ, Barry BK, Logan PA. Concurrent strength and endurance training: a review. Sports Med. 1999;28:413-27.
12. Marx JO, Ratamess NA, Nindl BC, et al. Low-volume circuit versus high-volume periodized resistance training in women. Med Sci Sports Exerc. 2001;33(4):635-43.
13. Mayhew JL, Ware JS, Johns RA, Bemben MG. Changes in upper body power following heavy-resistance strength training in college men. Int J Sports Med. 1997;18:516-20.
14. Moss BM, Refsnes PE, Abildgaard A, Nicolaysen K, Jensen J. Effects of maximal effort strength training with different loads on dynamic strength, cross-sectional area, load-power and load-velocity relationships. Eur J Appl Physiol Occup Physiol. 1997;75:193-9.
15. Mulligan SE, Fleck SJ, Gordon SE, Koziris LP, Triplett NT, Kraemer WJ. Influence of resistance exercise volume on serum growth hormone and cortisol concentrations in women. J Strength Cond Res. 1996;10:256-62.
16. Murphy AJ, Wilson GJ. The assessment of human dynamic muscular function: a comparison of isoinertial and isokinetic tests. J Sports Med Phys Fitness. 1996;36:169-77.
17. Newton RU, Häkkinen K, Häkkinen A, McCormick M, Volek J, Kraemer WJ. Mixed-methods resistance training increases power and strength of young and older men. Med Sci Sports Exerc. 2002;34(8):1367-75.
18. Newton RU, Kraemer WJ. Developing explosive muscular power: implications for a mixed methods training strategy. Strength Cond J. 1994;16:20-31.
19. Newton RU, Kraemer WJ, Häkkinen K. Effects of ballistic training on preseason preparation of elite volleyball players. Med Sci Sports Exerc. 1999;31(2):323-30.
20. Newton RU, Kraemer WJ, Häkkinen K, Humphries BJ, Murphy AJ. Kinematics, kinetics, and muscle activation during explosive upper body movements. J Appl Biomech. 1996;12:31-43.
21. Nindl BC, Harman EA, Marx JO, et al. Regional body composition changes in women after 6 months of periodized physical training. J Appl Physiol. 2000;88:2251-9.
22. Poliquin C. Five steps to increasing the effectiveness of your strength training program. Natl Strength Cond Assoc J. 1988;10:34-9.
23. Rhea MR. Determining the magnitude of treatment effects in strength training research through the use of the effect size. J Strength Cond Res. 2004;18:918-20.
24. Rhea MR, Ball SD, Phillips WT, Burkett LN. A comparison of linear and daily undulating periodized programs with equated volume and intensity for strength. J Strength Cond Res. 2002;16:250-5.
25. Schiotz MK, Potteiger JA, Huntsinger PG, Denmark DC. The short-term effects of periodized and constant-intensity training on body composition, strength, and performance. J Strength Cond Res. 1998;12:173-8.
26. Schlumberger A, Stec J, Schmidtbleicher D. Single- vs. multiple-set strength training in women. J Strength Cond Res. 2001;15:284-9.
27. Stone MH, O'Bryant HS, Schilling BK, et al. Periodization: effects of manipulating volume and intensity. Part 1. Strength Cond J. 1999;21:56-62.
28. Stone MH, Potteiger JA, Pierce KC, et al. Comparison of the effects of three different weight-training programs on the one repetition maximum squat. J Strength Cond Res. 2000;14:332-7.
29. Willoughby DS. The effects of mesocycle-length weight training programs involving periodization and partially equated volumes on upper and lower body strength. J Strength Cond Res. 1993;7:2-8.


©2009The American College of Sports Medicine