Analysis of Footstep/Stride Length from Gait Patterns of Dynamic Footprints as a Parameter for Biological Profiling—A Preliminary Study

Abstract

In forensic sciences, particularly in forensic anthropology and podiatry, assessing a person’s stature helps create a biological profile that allows for more accurate identification. Background/Objectives: When considering dynamic footprints as part of the gait pattern, certain parameters such as stride length, step length, gait width, and gait angle can be evaluated in relation to stature. The aim of this study was to assess footstep and stride length from the gait of dynamic footprints and determine if they correlate with stature and could be useful for biological profiling. Methods: Gait patterns from dynamic footprints and stature were determined in 114 females and 104 males aged 18 to 33 years. Results: All participants took the first step with their preferred foot, 56% with the right foot. Regarding step sequence, there were non-significant differences between the 4th and 5th footsteps in both sexes. Sex differences were significant in four of seven footsteps. Only a few steps significantly correlated in sequence with stature, and even these had low correlation coefficients (r = 0.295). In females, positive values of mean differences between actual and estimated stature predictions indicate that the equations tend to overestimate, whereas in a mixed sex group, most negative values of mean differences indicate underestimation. Conclusions: Given the weak correlations observed, footstep and stride length should not be considered reliable indicators for forensic stature estimation. These parameters are more suitable for biomechanical and anthropological research, while forensic applications should be considered supplementary and interpreted with caution.

1. Introduction

In addition to fingerprints and DNA, footprints found at the crime scene are another potential feature that can help identify through a biological profile [1], which includes the so-called Big Four—stature, age, sex, and ancestry [2]. In forensic science, particularly in forensic anthropology and forensic podiatry, estimation of stature using foot measurements [3,4,5,6], footprint measurements [7,8,9,10,11,12,13], and gait parameters [14,15,16,17] can play an important role in assessing a person’s biological profile to aid in the identification process, especially when other data are not available. Gait characteristics may be represented by shoeprints or footprints, which can be found at certain types of crime scenes and collected as evidence, depending on the surface, conditions, and actions of the perpetrator. Although footprints may contribute to forensic identification, their use for stature estimation remains in the investigative phase, requiring further validation before being considered a reliable forensic tool. The study of barefoot prints is not only important in developing countries, where the majority of the rural population prefers to walk barefoot due to socioeconomic and climatic reasons [1], but is also gaining importance in developed countries [1,9,18,19]. There are a few studies that examine footprints in terms of the mechanism of formation, in terms of static and dynamic footprints [1,9,18,19], and their relationship with stature and weight [7,8,9,12]. When considering dynamic footprints as part of the gait pattern, specific parameters such as stride length, footstep length, gait width, and gait angle can be evaluated. Few studies have attempted to document the relationship between stature and footstep/stride length [14,15,16,17,20,21].

The formation of dynamic footprints can be explained using the mechanism of walking. Gait is a biological characteristic of a person and gait pattern is the way or style in which a person usually walks. The gait cycle is divided into two phases: the time when the foot is on the ground (stance) and the time when the foot is off the ground (swing). When walking, our feet communicate with the ground/surface in a stereotypical way: heel strike, forefoot roll, and subsequent rebound with the distal forefoot and toes [22]. The arch of the foot plays an important role when walking, as it bears the entire weight of the body. When walking, we push the ground with our first toe and then transfer the weight to the second or third toe, while the fourth and fifth toes are less loaded. It is related to the dynamic nature of the medial arch of the foot arch and its more significant involvement in locomotion, as well as lower strength of ligaments and tendons and increased laxity of the lateral arch [9]. In the case of constant loading of the arch, it may drop and cause a flat foot, which appears on the impression of the foot in the form of an impression of the entire lower part of the foot [23].

Individuals move their bodies and limbs in very unique and repeatable patterns, and camera-based computer systems can be taught to recognize these patterns. It is possible to obtain some walking parameters from dynamic footprints, such as footstep length or footstep width. Stride length is the distance from one heel strike to the second heel strike of the same foot. Footstep width is the mediolateral distance between the right and left footprints [22,23,24].

The human locomotion system is unique in an evolutionary sense, as bipedalism required significant skeletal modifications, including foot structure and gait adaptation. The biomechanics of upright posture and walking are very different from the movement of a quadruped, so there had to be many modifications to the skeleton, including the legs. In the course of phylogenetic evolution, the human foot changed from a grasping organ to a supporting organ. The development of the double curvature on the foot meant the emergence of a longitudinal and transverse arch, which allow flexible locomotion and cushioning of the impacts that occur with each footstep, so that they are not transmitted in full intensity to the vital organs. Thus, the tread of the foot depends on the shape of the arch, and the elasticity of the arch allows the foot to adapt to the unevenness of the terrain, transmit the weight of the body, and ensure movement [25]. The footstep is the basic unit of terrestrial movement in bipeds. Like walking, the bipedal footstep can be divided into two parts: the support part (rest) when the limb is in contact with the mat, and the swing part (movement) when it is moving across the mat. Footsteps and strides are measured from the point of first contact of the foot (shoe) with the mat, which in normal walking is the heel. In running, when the heel bounces off the mat or not at all, the length of the footstep and stride is measured from the tips of the toes (or the front edge of the shoe) [20,26].

Loading the foot with one’s body weight creates a footprint—a footprint from which the individual’s stature, weight, age, sex, and ancestry can be estimated based on the selected measurements and morphological shape, allowing for biological profiling through the identification process [27]. Leaving multiple consecutive footprints involves continuously applied locomotion traces, which can be used to measure the length of a footstep and the stride length, and based on this, the relationship to the stature of the person who left each footprint can be determined [14,15,17,20,28].

In recent years, the use of artificial intelligence and computer vision in gait analysis has expanded significantly, offering new opportunities for both biometric and forensic applications. Several recent studies [29,30,31,32] have investigated gait attribute recognition using deep learning models and multi-angle datasets to improve identification accuracy and proposed frameworks for suspect tracking based on AI-powered gait features from video surveillance. Studies also emphasize the role of digital footprint analysis and motion coding in forensic and security scenarios, including through the integration of smart sensors and robotics. Although such approaches require large, high-quality video datasets and advanced infrastructure, they represent the developing area of forensic science. This emphasizes the need to continuously evaluate traditional methods, such as the footprint-based parameters used in this study, in this evolving technological context.

The aim of the present work is to analyze the relationship between stature and gait parameters (footstep and stride length) obtained from dynamic footprints and to assess whether they are reliable features for further estimation of stature with sufficient accuracy, which is used in forensics for biological profiling.

2. Materials and Methods

Measurements were made on 218 Slovak participants, of whom 114 were female and 104 were male and aged between 18 and 34 years, as research shows that aging is associated with changes in gait parameters, such as slower speed, shorter footstep length, and longer double support time [33]. All measurements were recorded using participant number, age, sex, date of birth, and date of examination. Decimal age was calculated [34] with a mean age of 22.73 years for males and 23.70 years for females. Participation was voluntary and based on written informed consent, which included details about the study, its objectives, methods, and expected participation. Participants were guaranteed that they could withdraw from the study without giving any reason. All procedures performed were in accordance with ethical standards, the Statute of the Local Ethics Committee for Research Involving Human Subjects at the Faculty of Natural Sciences of Comenius University in Bratislava, Slovakia (evaluated under the number ECH19004), as well as relevant legal regulations and the Declaration of Helsinki.

Inclusion criteria: Participants in the study were between 18 and 30 years old, were able to give informed consent, and had no lower limb abnormalities, deformities, or medical problems that could affect their ability to walk or stand.

Exclusion criteria: Subjects with spinal or lower limb deformities, diseases or surgical procedures, or foot abnormalities that could affect their locomotion were excluded. Pregnant women were also excluded from the study.

2.1. Anthropometric and Gait Measurements

Of the anthropometric parameters, only stature was measured for the present study. Measurements were made according to standard anthropometric procedures using a standard anthropometer—from the vertex to the floor when the subject was barefoot in the anatomical position and with the head in the Frankfurt plane—by a single investigator [35]. All participants were measured at approximately the same time in the morning to avoid diurnal variations in stature [36,37].

The procedure for obtaining dynamic bare footprints using the validated cream method in the context of gait analysis is described in detail in the previous study by Švábová et al. 2022 [8]. Briefly, seven footsteps were recorded for each participant in the form of dynamic footprints on paper. The terms “footstep” and “stride” used in this study are based on the definitions of previous studies [20,26] and may differ from the terminology used in other literature on gait analysis:

-

Footstep length was measured as a straight distance from the rearmost point of the heel of one footprint to the rearmost point of the heel of consecutive footprint (Figure 1A) [20,28]. The term “footstep” represents the single footstep taken by a participant and is used in the present study with the defined meaning.

-

Stride length was measured as the straight distance between the rearmost point of the heel of one footprint and the rearmost point of the heel of the following footprint on the ipsilateral side (Figure 1B) [20,28]. Thus, the stride represents a double step of a participant, and in the present study, the term “stride” is used with the defined meaning. Stride length was used in the present study following the meaning found in previous studies [14,15,17,20] in relation to stature.

While footstep length and stride length are related parameters, they provide distinct biomechanical insights. Therefore, both were analyzed separately to explore their individual contributions to stature estimation.

The study intentionally examined all recorded footsteps, including the initial ones, to confirm that the fourth and fifth footsteps represent optimal gait parameters. Unlike other studies that exclude early footsteps due to acceleration effects, our approach ensures that footstep patterns are not preselected, allowing for a more comprehensive analysis of gait dynamics.

2.2. Statistical Analysis

All data obtained were statistically analyzed using IBM SPSS Statistics Base for Windows, Version 20.0 (IBM Corp., Armonk, NY, USA). Normality testing was performed to examine the distribution of raw data; however, given the sample size, the assumption of normality in the sampling distribution is expected to hold due to the central limit theorem. Differences between the sexes were tested by the t-test for independent samples. An analysis of variance (general linear repeated-measures model) with multiple comparisons was used to test for differences between sequence of footsteps and laterality. An independent samples t-test was used to compare footstep parameters between sexes, as our primary goal was to assess mean differences rather than classification accuracy, which would typically be the focus of discriminant analysis. Sex was treated as a grouping variable to enable interpretation of gait parameters (footsteps and strides) between sexes (females and males). All analyses (descriptive, correlation, and regression) were conducted separately for males, females, and for the mixed sex group. Pearson correlation analysis was performed to determine the relationship between stature, footstep length, and stride length. Linear regression analysis (enter method and stepwise method) was used to develop predictive equations. Stepwise regression was employed as an exploratory tool to identify the most predictive gait parameters for stature estimation.

3. Results and Discussion

Table 1. Descriptive characteristics of footstep lengths, stride lengths, and stature with regard to sex.

Gait Parameters (in cm)	Females (N = 114)				Males (N = 104)
	Mean	SD	Min	Max	Mean	SD	Min	Max
Footstep lengths (FSL)
FSL 1	46.89	8.13	30.20	70.70	49.23	8.39	30.70	73.40
FSL 2	53.32	8.70	32.50	76.10	53.45	7.87	35.80	78.00
FSL 3	56.36	8.07	31.20	74.50	58.52	7.38	39.50	75.50
FSL 4	58.84	7.32	39.10	82.20	59.84	7.53	40.10	76.30
FSL 5	58.48	6.02	44.30	74.80	60.83	7.32	42.70	78.80
FSL 6	55.08	5.38	43.30	64.90	58.43	7.15	35.80	73.40
FSL 7	52.13	7.11	27.00	70.10	54.69	6.70	38.90	66.70
Stride lengths (SL)
SL 1	100.33	15.42	66.80	143.40	102.77	14.80	66.90	150.30
SL 2	109.76	15.85	67.10	149.00	111.88	14.10	84.20	152.70
SL 3	115.17	14.49	70.30	150.70	118.35	13.88	86.30	150.20
SL 4	116.46	13.34	76.80	152.60	120.71	13.77	90.20	151.00
SL 5	112.44	10.43	80.80	133.90	118.96	12.84	90.50	141.70
SL 6	107.53	9.92	86.10	134.50	112.23	11.81	71.90	137.20
Stature	166.86	6.04	152.10	186.30	180.33	6.98	161.00	198.40

shows the basic statistical characteristics of stature, footstep length (FSL), and stride length (SL) with regard to sex. In females, the fourth footstep (58.84 cm) reaches the highest value, while in males it is the fifth footstep (60.83 cm). In a group of 198 males, Jasuja et al. [15] found the longest third and fourth footsteps (73.9 cm). In both sexes in the studied sample, the first footstep is shorter than the others, being 46.89 cm in females and 49.23 cm in males. The study by Tripathy [38] showed in a sample of 209 males that 56% of the individuals had a longer first footstep than the other nine. In the studied sample of females, the values decrease from the fifth to the seventh footstep, and in males this tendency is observed from the sixth to the seventh footstep. Based on this observation, it can be concluded that the first footstep represents the so-called start footstep and the seventh or last footstep represents the so-called catch-up footstep, as they are shorter compared to the other footsteps. Another reason for the last two footsteps being shorter than the others could be due to the length of the paper (5 m) used to capture the dynamic footprints representing the gait pattern. However, this does not necessarily mean that they do not represent values for subjectively normal footsteps, since individuals may tend to take longer or shorter footsteps than is actually natural for them when taking other footsteps. It has been noted that the gait of some individuals is not regular and has an individual character that is subsequently reflected in the traces of locomotion left behind [20]. According to a study by Jasuja and Manjula [14], some taller individuals tend to take shorter but more frequent footsteps and shorter ones tend to take larger footsteps. Although some taller individuals took shorter but more frequent footsteps, the forensic applicability remains limited due to the high individual variability. This reinforces the necessity of using multiple biometric traits in forensic casework rather than relying solely on gait analysis.

More than 70% of the individuals of the mixed sex group as well as the females have footstep values in the range of 50 cm to 65 cm, and for the males this range is 55.10 cm to 70 cm. Only 4.72% of the individuals in the mixed sex group had a footstep length between 70.10 cm and 76.30 cm (

Table 2. Frequency of footstep range values in study sample with regard to sex.

Range of Footstep Lengths (cm)	Mixed Sex Group (N = 212)		Females (N = 109)		Males (N = 103)
	n	%	n	%	n	%
45.00–50.00	15	7.08	8	7.34	7	6.80
50.10–55.00	44	20.75	27	24.77	17	16.50
55.10–60.00	55	25.94	30	27.52	25	24.27
60.10–65.00	52	24.53	28	25.69	24	23.30
65.10–70.00	36	16.98	13	11.93	23	22.33
70.10–76.30	10	4.72	3	2.75	7	6.80
total	212	100	109	100	103	100

). In females, this percentage was lower (2.73%) than in males (6.93%) and at the same time identical to the length category of 45.00 to 50.00 cm. The number of individuals in each category in relation to the length of the footstep is shown in Figure 2. Based on the results in Table 2 and Figure 2, a higher proportion of males fall into the longest footstep length category (65.00–70.00 cm), while a greater proportion of females are represented in the 55.10–60.00 cm footstep length category. However, the most common footstep length interval for both sexes is 55.10–60.00 cm, which includes 27.27% of females and 24.75% males.

In the study by Tripathy [38], 75% of males were found to have a footstep length of between 55 and 70 cm. As shown in Table 2 and Figure 2, in our study a very similar proportion—71.28% of males—also fell within this range. Only 6.93% of our male participants had short footsteps (45.00–50.00 cm), which is similar to the 9% reported by Tripathy [38]. On the other hand, no male participant in our sample exceeded the 70 cm category, whereas Tripathy observed 2.8% of males with footstep lengths over 75 cm. These values confirm that our results are consistent with those of Tripathy, especially regarding the distribution in the 55–70 cm range.

3.1. Analysis of the Length of Footsteps and Strides in Relation to Sex, Sequence, and Laterality

Table 3. Number (N) of footsteps with respect to sequence with regard to laterality and sex.

Sequence of Footsteps	N Right	N Left	N Total	% Right	% Left
Mixed sex group
1	124	96	220	56.36	43.64
2	97	124	221	43.89	56.11
3	125	96	221	56.56	43.44
4	96	125	221	43.44	56.56
5	123	92	215	57.21	42.79
6	82	103	185	44.32	55.68
7	76	47	123	61.79	38.21
Females
1	63	52	115	54.78	45.22
2	53	63	116	45.69	54.31
3	64	52	116	55.17	44.83
4	52	64	116	44.83	55.17
5	63	50	113	55.75	44.25
6	45	50	95	47.37	52.63
7	39	26	65	60.00	40.00
Males
1	60	44	104	57.69	42.31
2	44	60	104	42.31	57.69
3	60	44	104	57.69	42.31
4	44	60	104	42.31	57.69
5	59	42	101	58.42	41.58
6	37	52	89	41.57	58.43
7	36	21	57	63.16	36.84

shows the distribution of footsteps across seven sequential footsteps during the walking trials, categorized by laterality (right or left foot) and sex (mixed group, females, and males). As participants started walking with their preferred foot to ensure a natural gait pattern, the number of contacts with the right and left foot during the gait sequence differed. For each footstep (from the 1st to the 7th), the number of right and left footsteps is given, together with the total number of observations and the corresponding percentages. In the mixed sex group, a slight preference for the right foot was observed at the initiation of gait, especially during the first footstep (56.36% right). With respect to sex, males showed a more pronounced tendency to use the right foot at the beginning of the gait (57.69% in the first footstep and up to 63.16% in the seventh footstep), while females showed a more balanced pattern between right and left footsteps, with only a slight dominance of the right foot in the first footsteps (54.78% in the first footstep).

Table 4. Comparison of footstep (FSL) and stride lengths (SL) with regard to sex and sequence.

Gait Parameters (in cm)	Females (N = 114)		Sequency Comparison	Males (N = 104)		Sequency Comparison	Sex Differences
	Mean	SD	p-Value	Mean	SD	p-Value	p-Value
FOOTSTEPS
FSL1	46.89	8.13	0.000	49.23	8.39	0.000	0.038
FSL2	53.35	8.74		53.45	7.87		0.904
FSL2	53.32	8.70	0.000	53.45	7.87	0.000
FSL3	56.36	8.07		58.52	7.38		0.042
FSL3	56.36	8.07	0.000	58.52	7.38	0.015
FSL4	58.84	7.32		59.84	7.53		0.318
FSL4	58.85	7.11	0.400	59.78	7.60	0.065
FSL5	58.48	6.02		60.83	7.32		0.012
FSL5	58.02	5.52	0.000	60.42	6.96	0.002
FSL6	55.20	5.30		58.43	7.15		<0.000
FSL6	55.35	4.84	0.001	58.40	5.68	0.000
FSL7	52.13	7.11		54.69	6.70		0.047
STRIDES
SL1	100.33	15.42	0.000	102.77	14.80	0.000	0.237
SL2	109.76	15.92		111.88	14.10		0.300
SL2	109.76	15.85	0.000	111.88	14.10	0.000
SL3	115.17	14.49		118.35	13.88		0.099
SL3	114.67	14.32	0.004	117.94	13.82	0.000
SL4	116.46	13.34		120.71	13.77		0.024
SL4	115.04	11.92	0.001	119.63	13.83	0.312
SL5	112.44	10.43		118.96	12.84		<0.000
SL5	111.74	9.76	0.000	117.56	10.81	0.000
SL6	107.53	9.92		112.23	11.81		0.020

shows the results of the multiple comparison for the footsteps and strides analyzed in relation to sex and sequence but independent of the foot laterality with which the gait was initiated.

Table 5. Comparison of footsteps (FSL) with regard to sex, laterality (right, left), and sequence (first to seventh).

Gait Parameter (in cm)	Females (N = 114)		Comparison	Males (N = 104)		Comparison	Sex Differences
	Mean	SD	p-Value	Mean	SD	p-Value	p-Value
FSL right 1	46.46	7.74	0.000	47.66	7.93	0.000	0.415
FSL left 2	52.43	8.51		52.45	7.80		0.965
FSL left 1	47.26	8.57	0.000	51.38	8.62	0.001	0.021
FSL right 2	54.75	8.71		54.82	7.83		0.857
FSL right 2	54.75	8.71	0.000	54.82	7.83	0.000
FSL left 3	57.84	8.53		59.96	7.82		0.211
FSL left 2	52.39	8.45	0.000	52.45	7.80	0.000
FSL right 3	55.25	7.48		57.46	6.92		0.083
FSL right 3	55.20	7.43	0.000	57.46	6.92	0.006
FSL left 4	57.44	7.31		59.29	7.78		0.173
FSL left 3	57.84	8.53	0.004	59.96	7.82	0.492
FSL right 4	60.52	6.89		60.59	7.19		0.960
FSL right 4	60.38	6.99	0.305	60.64	7.35	0.921
FSL left 5	59.84	5.58		60.81	10.69		0.578
FSL left 4	57.27	7.25	0.246	59.16	7.78	0.330
FSL right 5	56.24	7.72		59.82	7.37		0.010
FSL right 5	56.90	6.49	0.017	59.30	6.83	0.046
FSL left 6	54.04	7.52		57.94	7.78		0.012
FSL left 5	58.84	4.84	0.000	60.36	10.94	0.536
FSL right 6	55.10	4.61		59.12	6.18		0.001
FSL right 6	54.73	4.19	0.009	58.60	5.08	0.001
FSL left 7	51.07	5.65		52.86	6.98		0.336
FSL left 6	55.66	5.16	0.025	58.05	6.15	0.011
FSL right 7	52.95	7.76		54.17	9.33		0.538

then shows the results of the multiple comparison for the analyzed footsteps and strides in terms of sex and laterality of the foot with which the gait was initiated, together with the attention to the sequence. Obviously, it is not possible to compare two matching first footsteps of the same individual. Unlike previous studies [20,38] that discard initial footsteps due to acceleration effects, our study examined all footsteps to verify that the fourth and fifth footsteps represent stable gait characteristics. This approach ensures a more comprehensive understanding of footstep variability.

In males, the values of all evaluated footstep lengths are higher than in females (Table 4). However, these differences were found to be significant only for the lengths of the first, third, fifth, and sixth footsteps (p < 0.05). When comparing stride lengths, there were significant differences (p < 0.05) for the fourth, fifth, and sixth strides. These differences were higher for the first, fifth, and sixth footsteps than for the third footstep. Based on this finding and considering the insignificant sex differences (p > 0.05) for the length of the second and fourth footsteps, we can conclude that individuals tend to shorten (for males) or lengthen (for females) the footstep, which may not correspond to the normal footstep values for each sex. On the other hand, the same explanation can be given for the significance of the differences between sexes, since sex may not be a significant factor affecting the length of the footstep, which means that the dimensions of the footsteps that have significant dimensions may not represent the values of the normal footsteps. In several studies dealing with gait parameters, no differences were found between sexes because the group of subjects studied was male [14,15,20,38]. Significant sex differences in footstep length were confirmed in a study by Jansen et al. [39] that examined the effect of age on various gait parameters (using a treadmill) in a sample of 20 subjects aged 20–29 years and 20 subjects aged 60–69 years.

The results of the repeated measures tests (GLM for repeated measures) confirmed significant differences between the first through seventh footsteps and the first through sixth strides (p < 0.001). In addition, Table 5 shows the specific differences between the footsteps and the strides in relation to the sequence. Statistically non-significant differences (p > 0.05) were found between the fourth and fifth footsteps in males and females and between the fourth and fifth strides in males. Previous studies on footprints [9,40,41,42,43] also report that the mid-gait protocol should be followed when analyzing dynamic footprints, which confirms and supports the results of the present study. This confirms the previous statement regarding the footsteps, that the fourth and fifth footsteps should represent normal and stable values of the gait pattern, but at this stage the results are exploratory. According to the results, the other footsteps have a relatively irregular character, which is subsequently reflected in the character of walking/gait pattern as such. Several studies report five or more footsteps in their gait studies, but they do not report differences in the sequence of footsteps/strides [20,38]. The study by Jasuja et al. [15] found insignificant differences between the first five footsteps examined, which were subsequently included in the study, and the other footsteps were discarded.

Based on the results from the studied sample, the average footstep length (from the measurements of the fourth and fifth footstep, since the differences were not significant) for females was 58.66 ± 6.17 cm, for males 60.30 ± 6.90 cm, and in the mixed sex sample 59.41 ± 6.54 cm, confirming the normal distribution of the data (for the females p = 0.380, for the males p = 0.532, in the total sample p = 0.390). The value for footstep length seems to be higher in males than in females, but this difference is not statistically significant (p = 0.07).

The present results on the differences in the sequence of footsteps can also be understood as bilateral differences, with each footstep (1–7) simultaneously representing the right or left. Since everyone stepped with their preferred leg/foot, a comparison of laterality of the same footstep sequence was not possible in this way. Specific comparisons of the laterality of footsteps according to the preferred foot/lower limb with which the individual performed, along with the results of differences between sexes in these parameters, are shown in Table 3. Thus, for further explanation, the first right footstep was compared to the second left footstep and vice versa for the remaining footsteps. Also in this case, no significant differences (p > 0.05) were confirmed between the fourth and fifth footsteps and for males for the third/fourth and fifth/sixth footsteps. In this case, the gait of the male subjects in the studied sample seems to be less dynamic and relatively unstable. The differences between the sexes were also significant for the first, fifth, and sixth footsteps. The observed instability in male gait could be influenced by the limited number of steps analyzed. Future research should assess longer gait sequences to determine whether footstep length variation persists over extended walking distances.

The study by Jasuja and Manjula [14] presents results for right and left footsteps but does not indicate differences. The study by Straus [20] found non-significant differences between right and left footsteps during subjective normal walking on hard surfaces without weight load and other influences. Significant differences were found between right and left footsteps when a person walked with a load in the right hand and a load weight of more than 15 or 17 kg. When a person walked with a load of more than 15 kg, the study confirmed a change in the dynamic walking stereotype; at a lower load, the study did not find a significant change in the walking stereotype.

3.2. The Association Between Length Parameters of Footstep/Stride and Stature

The expected relationship between stature and the length measures of footsteps and stride was confirmed only in females for the first, second, and third footsteps and strides and in the sex-mixed group for the first, third, fifth, and sixth footsteps and first through fifth strides (

Table 6. Pearson’s correlation coefficient (r) between stature and footstep/stride length in males, females, and mixed sex group.

	Females	Males	Mixed Sex Group		Females	Males	Mixed Sex Group
	N = 114	N = 104	N = 218		N = 114	N = 104	N = 218
Footstep length (FSL)				Stride length (SL)
FSL1-r	0.236	0.140	0.229	SL1-r	0.272	0.094	0.184
p-value	0.012	0.155	0.00	p-value	0.004	0.342	0.007
FSL2-r	0.269	0.026	0.110	SL2-r	0.302	0.105	0.192
p-value	0.004	0.791	0.107	p-value	0.001	0.290	0.004
FSL3-r	0.295	0.160	0.256	SL3-r	0.248	0.128	0.209
p-value	0.001	0.104	0.000	p-value	0.008	0.195	0.002
FSL4-r	0.169	0.079	0.133	SL4-r	0.091	0.105	0.180
p-value	0.072	0.424	0.049	p-value	0.343	0.294	0.009
FSL5-r	0.033	0.153	0.197	SL5-r	0.141	0.135	0.291
p-value	0.732	0.127	0.004	p-value	0.179	0.207	<0.000
FSL6-r	0.144	0.068	0.256	SL6-r	0.140	−0.072	0.176
p-value	0.174	0.524	0.001	p-value	0.273	0.596	0.054
FSL7-r	0.118	−0.172	0.122	FSL (mean)-r	0.072	0.121	0.159
p-value	0.358	0.209	0.188	p-value	0.457	0.230	0.022

Legend: N—number of individuals, FSL—footstep length, SL—stride length, r—correlation coefficient, p-value—significance value.

). There was no statistically significant correlation in males, which may be due to the greater variability in footstep and stride lengths (gait parameters analyzed) and also because individual walking dynamics appeared to be less consistent across footstep sequences, possibly due to greater diversity in body composition or gait style. In addition, previous studies [14,15,17,20] have reported that taller individuals do not necessarily take longer footsteps, and this effect may be more pronounced in males, reducing the strength of the correlation between stature and gait length.

For females, the third footstep (r = 0.295) and the second stride (r = 0.302) were most strongly correlated with stature, and for the mixed sex group, the third and seventh footsteps (r = 0.256) and fifth stride (r = 0.291) were. In males, no significant correlation was found between stature and footsteps or strides. A negative correlation coefficient is observed for the last footstep and stride length, which could be due to the fact that the last footstep is the so-called catching up step and the walking surface is limited, and thus they are not the actual and representative values of the normal/characteristic gait pattern. Although the first and last footsteps show differences from mid-gait footsteps, excluding them completely might not always be appropriate, depending on the forensic context of footprint evidence at crime scenes.

The fourth and fifth footsteps are supposed to represent the optimal footstep length, and, at the same time, a stable footstep sequence based on insignificant differences in the sequence (Table 3 and Table 4), but they are not significantly correlated in both males and females. In the mixed sex group, it is significant only for the length of the fifth footstep (r = 0.197), but the correlation is low.

Table 6 also shows the correlation results for the average footstep length value (FSL mean) calculated from the fourth and fifth footsteps, since the order is not significant. A significant correlation with stature was confirmed only in mixed sex group. However, the value of the correlation coefficient of mean footstep length with stature (r = 0.159) is the lowest in the mixed group compared to other significant footstep length values in the mixed sex group.

A statistically significant relationship between stature and length measures of footsteps and strides was confirmed only in females (first, second, third footstep and stride) and in the mixed sex group (first, third, fifth, sixth footstep and up to the fifth stride). There were no significant correlations in males. Based on this, equations were created to estimate an individual’s stature using only footstep length and stride length where significant correlations with stature were found. The study by Kanchan et al. [17] focused on observing the correlation between footstep length and stature and lower limb length and found few positive correlations between footstep length and stature in females, and lower limb length could not be considered an important factor in determining the footstep length.

Table 7. Simple linear regression equations for estimating stature (in cm) from gait length parameters in females and in a mixed sex group.

Equations Footsteps	SEE	R2	Fa	Equations Strides	SEE	R2	Fa
Females (N = 114)				Females (N = 114)
158.590 + 0.176 × FSL1	5.916	0.056	6.533	156.097 + 0.107 × SL1	5.857	0.074	8.893
156.917 + 0.187 × FSL2	5.843	0.072	8.721	154.244 + 0.115 × SL2	5.784	0.091	11.214
154.412 + 0.221 × FSL3	5.796	0.087	10.686	154.973 + 0.103 × SL3	5.878	0.061	7.316
Mixed sex group (N = 218)				Mixed sex group (N = 218)
160.912 + 0.258 × FSL1	9.150	0.052	11.890	161.754 + 0.114 × SL1	9.240	0.034	7.509
155.689 + 0.307 × FSL3	9.069	0.065	15.113	160.029 + 0.120 × SL2	9.206	0.034	8.299
157.167 + 0.271 × FSL5	9.163	0.039	8.340	157.252 + 0.137 × SL3	9.173	0.044	9.898
151.575 + 0.383 × FSL6	9.481	0.065	12.446	158.662 + 0.123 × SL4	9.218	0.032	7.023
159.942 + 0.225 × FSL	9.227	0.025	5.345	146.104 + 0.235 × SL5	9.360	0.085	16.636

Legend: N—number of individuals, SEE—standard error of determination, R²—coefficient of determination, F^a—F ANOVA statistic, FSL—footstep length, SL—stride length.

shows the equations for footsteps and strides separately for the female and mixed sex groups, as well as the equation for footstep length determined from the fourth and fifth footsteps, based on the above results. The reported values of Fa confirm the significance of each model (p < 0.05). Table 7 also shows the standard errors of estimation (SEEs), which determine the deviation of calculated stature from the actual value of individual stature, i.e., the lower the value of SEE is, the more accurate the given parameter is in estimating individual stature. For females, these values are about the same for footsteps and strides, the lowest value is for the third footstep (SEE = ±5.796 cm) and the second stride (SEE = ±5.784 cm). For these two parameters, females also had the highest correlation coefficients compared to the other footsteps and strides. In the mixed sex group, the values of SEE are higher, so the predictive value of the footstep and strides in the sex-independent sample is less accurate than in the female group. In the mixed sex group, the lowest SEE values were observed for the third footstep (±9.069 cm) and the third stride (±9.173 cm). Nevertheless, the regression models show low R² values and a relatively high SEE, indicating that only a small part of the variability in stature is explained by gait parameters. Therefore, these models should be considered as exploratory models and not as reliable predictive tools.

The stepwise multiple linear regression method was used to determine the footstep and stride lengths that are theoretically most important and necessary for estimating an individual’s stature (

Table 8. Equations of multiple linear regression using the stepwise method to estimate stature (in cm) from gait length parameters in females and in a mixed sex group.

Equations Footsteps (FSL) and Strides (SL)	SEE	R2	Fa
Females (N = 114)
154.455 + 0.220 × FSL3	5.819	0.086	10.476
154.218 + 0.115 × SL2	5.803	0.091	11.139
Mixed sex group (n = 218)
153.175 + 0.360 × FSL3	9.453	0.069	13.089
144.770 + 0.262 × FSL3 + 0.246 * FSL6	9.371	0.091	8.709
145.881 + 0.237 × SL5	9.376	0.086	16.828

Legend: SEE—standard error of determination, R²—coefficient of determination, F^a—F ANOVA statistic.

). For females, it is the length of the third footstep and the second stride; for the mixed sex group, it is the third footstep, the combination of the third and sixth footsteps, and the fifth stride. SEE values are similar to SEE in simple linear regression equations. It is lowest for footstep in females (SEE = ±5.803 cm) and for the combination of third and sixth footstep in the mixed sex group (SEE = ±9.371 cm). In these cases, the same applies as for the simple regression equation above and only a small part of the variability in stature is explained by these gait parameters and should not be used as a reliable tool for predicting stature in forensics.

Despite the low predictive power of the regression equations, the validity of the calculated individual regression formulas was tested by calculating the mean difference between actual and estimated stature for each gait parameter for the females and the mixed sex group for each of the 40 randomly selected participants (

Table 9. Comparison of actual stature and estimated stature from length parameter of gait in 40 randomly selected participants for females and mixed sex group.

Stature (cm)
Females		Mixed Sex Group
Variable	Mean Differences	Variable	Mean Differences
FSL1	0.95	FSL1	−1.23
FSL2	0.48	FSL3	−0.77
FSL3	0.42	FSL5	−0.53
SL1	0.55	FSL6	0.10
SL2	0.34	SL1	−0.78
SL3	0.76	SL2	−0.55
		SL3	−0.69
		SL4	−0.50
		SL5	−0.35

Legend: FSL—footstep length, SL—stride length.

).

In the equations for stature, the mean difference was smallest for the second stride length (+0.34) in females and for the sixth footstep length (+0.10) in the mixed sex group. Negative values indicate that the equation tends to overestimate, whereas positive values indicate underestimation. The differences between estimated and actual stature for each prediction equation are smaller than the values of SEE. The close approximation is explained by the fact that the regression equations were calculated using measures of central tendency. The low accuracy of the present regression models can also be explained by low values of the coefficient of determination. In our study, we found that the footsteps or strides explained only 3.2% to 9.1% of the variability in stature (Table 7 and Table 8).

From the above results it can be concluded that the footstep length does not seem to be a stable variable, it is individually specific and influenced by several factors. For this reason, it seems to be an unreliable variable when estimating the stature of an individual. The study [44] came to a similar conclusion that stride length is not a reliable and appropriate variable for estimating stature, but this could have been due to the small number of subjects, as in our case. Despite the non-significant correlation in the present study in males, there are studies that found a positive correlation between stature and footstep length or stride length [14,15,17,20]. A negative correlation between footstep length and stature is reported by [39] and [45]. These results also indicate what [14] claims, that taller individuals tend to take more frequent and shorter footsteps, while shorter individuals tend to take longer footsteps.

In summary, differences between sexes in footstep length and stride length were significant for the first, third, fifth, and sixth footsteps and for the fourth, fifth, and sixth strides. Insignificant differences indicate the possibility of shortening or lengthening of the footstep, which would imply that true footsteps with significant differences between sexes represent the characteristic footstep length for each sex. Testing for differences between the sequence of footsteps and strides revealed non-significant differences only between the fourth and fifth footsteps, from which a mean was subsequently calculated. This finding confirms a stable and optimal footstep length value, which can be attributed to the stabilization and maintenance of the walking stereotype [46,47]. Accordingly, the first and last footsteps can be considered as so-called starting and catching up steps, which do not necessarily represent subjectively normal footstep length values due to the tendency to shorten or lengthen the respective footsteps. The observed differences in footstep sequence also represent differences between right and left footsteps, but the comparison of bilaterality with respect to footstep sequence was not possible because the subjects walked with the preferred leg/foot. A significant relationship between stature and length of footstep and stride was confirmed in females for some of the FSLs/SLs and in the mixed sex group. From a practical point of view, it would be optimal to calculate an average and stable footstep length that represents a natural walking style [46,47] and could be used to estimate stature, since in real crime scene scenarios footprints are not necessarily left as an optimal gait pattern with few footsteps. Since statistically significant differences between footsteps were found in the study sample, only the mean values of footstep calculated from the length of the fourth and fifth footsteps, for which no significant differences were found, could be informative for real-life forensic scenarios. However, the length of these footsteps correlated with stature only in the mixed sex group (when sex was taken into account, the correlation was not significant). The low accuracy of the developed regression equations for stature estimation can be seen in the high values of the mean differences between actual and predicted stature and as well as high values of the SEEs. In view of the above, it was confirmed that the footstep and stride are a relatively dynamic, unstable, and individually specific variable that is not accurate enough to be used for biological profiling. Based on the results obtained, gait seems to be influenced by several factors that affect the dynamic walking stereotype. In other words, not every person has a steady gait, and individuals with a larger body size may tend to take shorter footsteps, while conversely, small individuals may take longer footsteps.

Although stature estimation from dynamic footprints may not be a standalone forensic tool, it could serve as supplementary evidence in cases where other biometric data are incomplete or unavailable. It is important to note that the sample is limited to healthy young adults aged 18–33 from Slovakia and therefore the findings should not be generalized to other ethnic/ancestry, age, or demographic groups.

The results indicate that stature estimation from footstep and stride length lacks sufficient accuracy for forensic applications. The high degree of individual variability in gait parameters suggests that these measurements should not be relied upon for criminal investigations but may still contribute to academic research on human locomotion.

3.3. Study Limits and Further Recommendations

This study has some limitations that should be considered when interpreting the results. First of all, the analysis was carried out solely on the basis of barefoot footprints. While this approach ensures consistency in the measurement of anatomical features and footstep/stride length, it does not fully reflect real-life conditions at crime scenes where footprints are rarely left by bare feet (shoeprints are more common, especially in highly developed countries). Future research should therefore consider the inclusion of footprints made while wearing shoes and investigate how different types of footwear affect the relationship between stride length and stature. This would improve the forensic applicability and validity of footprint-based estimates. The study of other factors such as age, lower limb length, different environmental conditions, different surfaces—snow, clay—body weight or constitution of the participants, and individuals with different loads is beyond the scope of the study. These factors could affect the length of footsteps and strides, so further research should be conducted on these factors. Furthermore, most of the participants in this study were young adults, as aging is associated with changes in gait parameters, such as slower speed, shorter footstep length, and longer double support time [33]. It was beyond the scope of our study to investigate individuals with advanced age; however, it might be an interesting idea to include subjects with advanced age for further investigation. It is also important to note that the equations developed are population specific and cannot be generalized to other populations worldwide. Sex-specific analyses were conducted to explore potential biomechanical differences in gait, which may not be fully accounted for by including sex as a covariate in a single model. While this approach allowed for a more detailed examination of sex-based gait characteristics, it also reduced statistical power due to smaller sample sizes. Future studies should consider analyzing sex differences using interaction terms in regression models to retain statistical power while capturing potential differential effects. Additionally, stepwise regression was employed as an exploratory tool to identify the most predictive gait parameters for stature estimation. However, stepwise methods are known to have limitations, including overfitting and instability in model selection. As a result, the findings should be interpreted with caution, and confirmatory analyses using alternative statistical approaches are recommended for future research to enhance the robustness of stature estimation models based on gait parameters.

4. Conclusions

In summary, to the best of our knowledge, this is one of the first studies in Europe to examine and provide important conclusions for the basic knowledge of the use of gait parameters from footprints. These conclusions should not be generalized to other ethnic, age, or demographic groups as the sample is limited to healthy young adults aged 18–33 from Slovakia. However, given the low correlation coefficients and high estimation errors, footstep and stride length should not be considered reliable tools for stature estimation in forensic cases. These parameters are therefore better suited for biomechanical and anthropological research, and any indication of their forensic applicability should be considered supplementary. Future research should focus on integrating gait analysis with additional forensic markers to enhance identification accuracy. Based on our results, the low accuracy of the regression equations confirms that gait length parameters from footprints are not precise and robust enough to be used in biological profiling for stature estimation.