OPINION
BAKER, Judge.
Appellant-plaintiff Steven Lytle appeals the trial court’s order granting appellee-de-fendant Ford Motor Company’s (Ford) motion for summary judgment. Specifically, Lytle argues that the trial court erred as [467]*467follows: 1) excluding evidence of Kyong Ly-tle’s seat belt use; 2) concluding that Lytle had abandoned his theory regarding inadvertent release; and 3) excluding the testimony of Lytle’s expert witnesses.
FACTS1
On August 31, 1987, Lytle, his wife Kyong and their daughter Michelle were riding in their 1987 Ford Ranger pickup truck when it was struck by another vehicle. The force of the collision caused their truck to skid and roll over several times. Although Lytle contends that Kyong was wearing a seat belt at the time of the accident, she was thrown from the truck. As a result, Kyong suffered permanent brain damage, while Lytle and Michelle, who were restrained by their seat belts, incurred only minor injuries.
In August of 1989, Lytle filed a complaint against Ford, alleging that Kyong’s enhanced injuries were caused by a design defect in Ford’s seat belts. Specifically, Lytle alleged that the seat belt buckle inertially released as a result of the acceleration forces which occurred during the accident. In the alternative, Lytle alleged that the improper placement of the seat belt buckles combined with the ease with which Kyong’s buckle could be released, caused it to inadvertently release when it came in contact with either Michelle’s body or clothing, or her buckle. In response, Ford filed an answer denying Ly-tle’s allegations, contending that its seat belt design was not defective and that Kyong was not wearing her seat, belt at the time of the accident.
On August 23,1996, Ford filed a motion in limine seeking to exclude evidence of any design defects other than those relating to inadvertent release, inertial release or defects in the passenger door. Record at volume 8: page 1858. On September 19, 1996, during the hearing on Ford’s motion in li-mine, Lytle’s attorney informed the court that, “the only issue concerns the buckle and in that regard its [sic] simply the design of the buckle, the selection of this particular buckle compared to other safer alternative designs, and the failure to test the buckles.” R. at 39:9057. Thereafter,, the trial court granted Ford’s motion, concluding that the only issues remaining concerned the buckle’s design and selection and Ford’s failure to properly test the buckle. R. at 13:2889.
On August 23, '1996,' Ford also filed a motion in limine seeking to exclude the testimony of Lytle’s expert witnesses, Billy Peterson and John Marcosky. Specifically, Ford contended that Peterson’s testimony regarding inertial release was not scientifically reliable and would not assist the trier of fact. Ford further argued that any probative value that the testimony would provide would be substantially outweighed by the prejudice to Ford. Additionally, Ford contended that Marcosky’s testimony regarding inertial and inadvertent release “was not based on reliable analysis or knowledge and would not assist the trier' of fact.” R. at 8:1879. In September of 1996, after a hearing on the motion in limine, the trial court entered an order excluding Marcosky’s testimony because Lytle failed to demonstrate that the testimony was based upon any particular skill, knowledge, experience or expertise or that it would assist the jury. R. at 13:2891-92. The trial court also excluded Peterson’s testimony regarding inertial release because he could not show that the forces and circumstances which were present during his pendulum tests2 and which permitted the seat belts.to inertially release, were sufficiently similar to the forces and circumstances which are present in a “real world” .accident, or which were present during the. Lytle’s accident. R. at 13:2926-28.
Following the court’s ruling, during an offer of proof, Peterson testified that his pen[468]*468dulum tests demonstrated that inertial release ean occur at less than peak acceleration of between 40-60 g forces.3 Therefore, he argued, because Ford’s tests had demonstrated that buckles could be exposed to forces of 40-60 g’s during a real world accident, his tests proved that a seat belt could inertially release during a real world accident. However, the court again ruled that his testimony was inadmissible, finding that Lytle had failed to establish that the forces present during Peterson’s tests were similar enough to forces which are present in a real world accident.
On October 7,1996, Ford filed a motion for summary judgment, alleging that, without testimony from Marcosky and Peterson, Ly-tle could not establish a genuine issue of material fact regarding design. defect and causation. Thereafter, Lytle filed a motion in opposition to summary judgment, in which he argued that the testimony of his expert witnesses was admissible. Nevertheless, the trial court granted Ford’s motion for summary judgment and incorporated its previous rulings excluding Peterson’s testimony. The court also found that Lytle had abandoned his claim for inadvertent release, that Marco-sky’s and Peterson’s testimony regarding inertial release was inadmissible and that, without expert testimony, Lytle presented no genuine issue of material fact regarding defect, a safer alternative design or causation. R. at 38:9040-48. Lytle now appeals.
DISCUSSION AND DECISION
I. Standard of Review
Summary judgment is appropriate where no genuine issues of material fact exist and the moving party is entitled to judgment as a matter of law. Hottinger v. Trugreen Corp., 665 N.E.2d 593, 595 (Ind.Ct.App.1996), trans. denied. A trial court’s grant of summary judgment is “clothed with a presumption of validity,” and the appellant bears the burden of demonstrating that the trial court erred. Id. However, we must carefully scrutinize the trial court’s decision to ensure that the non-prevailing party is not improp-eriy denied his day in court. Id. Further, where expert testimony is advanced to establish causation, summary judgment is properly entered in favor of the defendant where that testimony fails to meet the admissibility requirements of Ind. Evidence Rule 702. Id.
II. Evidence of Kyong’s Seat Belt Use .
Lytle first contends that the trial court erred by failing to consider evidence that he offered in support of his motion in opposition to Ford’s summary judgment motion. Specifically, he argues that the court improperly excluded all evidence of Kyong’s seat belt use, including Lytlé’s deposition testimony that he saw Kyong put her seat belt on before the accident, Kyong’s mother’s testimony that Kyong always wore her seat belt and Marcosky’s affidavit, in which he stated that the marks on the seat belt demonstrate that Kyong was wearing her seat belt at the time of the accident.
In its order granting summary judgment, the trial court stated:
Purported evidence of seat belt use is submitted in support of the' opposition to the Motion for Summary Judgment. However, as it is submitted to the court it is hearsay. It is not offered through the testimony of any person or by affidavit and as it stands before the court in its present form it would not be admissible at trial therefore [sic] it should be disregarded.
R. at 38:9042. Thus, Lytle contends that the trial court disregarded Lytle and Kyong’s mother’s testimony and, Marcosky’s affidavit. However, we conclude that the court merely disregarded evidence which would not be admissible at trial. For example, in his brief in support of summary judgment, Lytle asserted that Michelle had told a police officer that Kyong was wearing her seat belt at the time of the accident. Evidence of Michelle’s statement, however, was not submitted through an affidavit, a police report or testimony from Michelle or the officer. Therefore, as it was presented, the statement was hearsay. See Ind. Evidence Rule 801(c) [469]*469(“Hearsay” is a statement, other than one made by the declarant while testifying at the trial or hearing, offered in evidence to prove the truth of the matter asserted.). As a result, the trial court properly disregarded the statement. See Kline v. Business Press, Inc., 516 N.E.2d 88, 91 (Ind.Ct.App.1987) (in ruling on motion for summary judgment, court properly disregarded inadmissible hearsay), trans. denied.
Although the trial court properly disregarded hearsay, we cannot conclude from the trial court’s statement that the court also excluded evidence of Kyong’s seat belt use which was offered through Lytle’s and Kyong’s mother’s deposition testimony and Marcosky’s affidavit, thus, precluding summary judgment on that basis. As a result, for purposes of this appeal, we will presume that Kyong was wearing her seat belt at the time of the accident.
III. Abandonment of Inadvertent Release Theory
Next, Lytle contends that the trial court erred by determining that, during the hearing on Ford’s motion in limine, he abandoned his theory that the seat belt buckle had inadvertently released. In response, Ford argues that, because Lytle’s inadvertent release theory involved only the location of the seat belt buckles and not the design, Lytle’s attorney’s statement, that the only issue remaining involved “the design ... selection ... [and] the failure to test the buckles,” constituted an abandonment of this theory. R. at 89:9056-57. We disagree.
The record reveals that Lytle’s inadvertent release theory is based, not only upon the improper location of the seat belt buckles, but also the relative ease with which Kyong’s buckle could be released as compared to other buckles. R. at 36:8424. Therefore, despite Ford’s contention to the contrary, Lytle’s theory of inadvertent release is not inconsistent with his attorney’s statement about the remaining issues. Furthermore, the record reveals that after Ly-tle’s attorney made this statement, Ford and Lytle continued to argue regarding the admissibility of Marcosky’s testimony on inadvertent release. R. at 39:9094r-95. Thus, it is apparent that neither party considered the theory to have been abandoned. In light of these circumstances and the strong public policy in favor of settling a case on its merits, we conclude that the trial court erred by determining that Lytle abandoned his theory of inadvertent release.4
IV. Expert Testimony
Next, Lytle contends that the trial court erred by excluding Peterson’s and Marco-sky’s expert testimony. Specifically, he contends that the court erred as follows: 1) excluding Marcosky’s testimony regarding inadvertent and inertial release; 2) excluding Peterson’s testimony regarding inertial release; and 3) failing to consider Peterson’s and Marcosky’s credentials in determining the reliability of their testimony.
The admissibility of an expert’s testimony is governed by Evid. R. 702 which provides as follows:
(a) If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.
(b) Expert scientific testimony is admissible only if the court is satisfied that the scientific principles upon which the expert testimony rests are reliable.
Thus, where an expert’s testimony is based upon the expert’s skill or experience rather than on the application of scientific principles, the proponent of the testimony must only demonstrate that the subject matter is related to some field beyond the knowledge of lay persons and the witness possesses [470]*470sufficient skill, knowledge or experience in the field to assist the trier of fact to understand the evidence or to determine a fact in issue. Evid. R. 702(a); Corbin v. State, 563 N.E.2d 86, 92-93 (Ind.1990). However, when the expert’s testimony is based upon scientific principles, the proponent of the testimony must also establish that the scientific principles upon which the testimony rests are reliable. Evid. R. 702(b).
Furthermore, the determination of the admissibility of all expert testimony is a matter within the sound discretion of the trial court. Buzzard v. State, 669 N.E.2d 996, 999 (Ind.Ct.App.1996). We will reverse only for an abuse of that discretion. Id.
A. Marcosky’s Testimony
Lytle first contends that the trial court erred by excluding Marcosky’s testimony regarding inertial and inadvertent release. Specifically, he contends that Marcosky’s testimony is admissible because it is based on his inspections of Kyong’s seat belt and the configuration of the seat belts in the Lytle vehicle, his pendulum tests and his “engineering knowledge, training and experience.” Appellant’s Brief at 32.
In his affidavit, Marcosky stated that, because the evidence revealed that Kyong was wearing her seat belt at the time of the accident, he believed that her seat belt either inertially or inadvertently released. R. at 36:8424. Marcosky’s deposition reveals that his opinion that Kyong’s buckle could have inadvertently released was based on his observations of Kyong’s buckle and the configuration of the seat belts. He contends that from these observations, he was able to conclude that Kyong’s buckle could be released with relatively little force and that the improper configuration of the seat belt buckles made it possible for Kyong’s buckle to come into contact with either Michelle or Michelle’s buckle. R. at 36:8424. Marcosky’s deposition also reveals that his opinion that Kyong’s seat belt may have inertially released was based upon his pendulum tests, in which he was able to cause a buckle to inertially release by hitting the back of it with his hand or a hammer.
In excluding Marcosky’s testimony, the trial court stated as follows:
[I]t does not appear to the court that Marcosky’s testimony is scientifically based or that there is a scientific foundation. His testimony does not ... appear to be helpful to the jury in that it does not require any particular, skill, knowledge, experience, or expertise to give the opinion or to give the evidence that Marcosky is anticipated to give. It appears any laymen [sic] could come up with [a] similar opinion given the knowledge of the circumstances.
R. at 13:2891-92.
First, we determine whether the trial court erred by excluding Marcosky’s testimony regarding inadvertent release. As noted above, Marcosky’s testimony regarding inadvertent release was based primarily upon his observations of Kyong’s buckle, the configuration of the buckles in the Lytle vehicle, his knowledge and his experience, rather than upon any scientific principles. Thus, Lytle was only required to establish that Marco-sky’s testimony was beyond the knowledge of lay persons and that he possessed sufficient skill, knowledge or experience that would assist the trier of fact. Corbin, 563 N.E.2d at 86.
Given Marcosky’s credentials, there is no doubt that he qualifies as an expert. However, Lytle has failed to demonstrate that Marcosky has performed any tests to determine that Kyong’s buckle could have been released with relatively little force as compared to other buckle designs. The record also reveals that Marcosky failed to perform any tests using surrogates who were the same size as Kyong and Michelle to determine whether the buckles could have come into contact with one another during an accident with enough force to cause inadvertent release. Finally, although Lytle contends that Marcosky’s opinion is also based upon his previous engineering knowledge, training and experience, Lytle has failed to demonstrate that Marcosky possessed any specialized knowledge, training or experience regarding the amount of forces generally required to release a buckle or proper seat belt configurations. Therefore, Lytle has failed to demonstrate that Marcosky possessed [471]*471skill, knowledge or experience that would assist the trier of fact to understand the evidence or to determine whether Kyong’s belt inadvertently released. In light of these circumstances, we cannot conclude that the trial court abused its discretion by excluding Marcosky’s testimony regarding inadvertent release.
Next, we determine whether the court erred by excluding Marcosky’s testimony regarding inertial release. The record reveals that the theory of inertial release is based upon complex scientific principles. Therefore, Lytle was required to establish that the scientific principles upon which Mareosky based his testimony were rehable. Evid. R. 702(b).
The record reveals that Marcosky’s opinion that Kyong’s buckle could have inertially released was based solely on the results of his pendulum tests,5 which were limited to hitting the back of a suspended buckle with his palm or a small hammer to determine if it would release. R. at 15:3349-50, 3352, 3357. This type of under evolved “pendulum” test has repeatedly been found to be unreliable by the National Highway Traffic Safety Administration (NHTSA) and the Society of Automotive Engineers (SAE) because it fails to take into consideration, among other things, web tension. Web tension increases as the seat belt occupant is thrown into the webbing of a seat belt during a roll over accident. R. at 9:2000-02, 1932, 1937, 1947. Previous tests have established, and Peterson and Mareosky do not dispute, that as web tension increases, the amount of g forces necessary to inertially release the buckle also increase. R. at 9:1940,1947. For example, a buckle that would inertially release at 60 g’s with no web tension, may not release at 200 g’s if there is even minimal web tension. R. at 9:1919. Thus, the factor that Marcosky’s tests ignore, web tension, is the very factor that has been shown to prevent a seat belt from inertially releasing during real world accidents. Because Marcosky’s testimony was based solely upon his pendulum tests and because the uncontroverted evidence reveals that these tests are unreliable, Lytle has failed to demonstrate .that Marcosky’s opinion is based on reliable scientific principles.6
B. Peterson’s Testimony
Next, Lytle contends that the trial court erred by excluding Peterson’s testimony regarding inertial release. Specifically, Lytle contends that the court erred by determining that Lytle failed to show that the scientific principles upon which Peterson based his opinion, that inertial release can occur in real world accidents, were reliable.7
Ah noted above, the theory of inertial release involves complex scientific principles. Thus, Lytle had the burden to prove the reliability of the scientific principles and tests upon which Peterson’s testimony was based. See McGrew v. State, 682 [472]*472N.E.2d 1289, 1290 (Ind.1997) (proponent of scientific expert testimony has burden to prove reliability). Scientific knowledge is more than subjective belief or unsupported speculation. Hottinger, 665 N.E.2d at 596. Expert testimony must be supported by appropriate validation or “good grounds” based on what is known, establishing a standard of evidentiary reliability. Id. Scientific validity for one purpose is not necessarily scientific validity for other unrelated purposes. Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 591, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). As such, proffered scientific evidence requires a valid scientific connection to the pertinent inquiry as a precondition to admissibility. Id. at 591-592, 113 S.Ct. 2786. Any inference or assertion must be derived by the scientific method. Hottinger, 665 N.E.2d at 596. If this nexus is not sufficiently established, the trial court may conclude that the analytical gap between the data and the opinion proffered is too great. General Electric Co. v. Joiner, 522 U.S. -, 118 S.Ct. 512, 519, 139 L.Ed.2d 508 (1997).
During Lytle’s offer of proof, Peterson testified regarding the procedures and results of the pendulum tests he performed in an attempt to prove that inertial release can occur in a real world accident. During his testimony, Peterson explained that the results of tests conducted by other experts revealed that, while occupying seat belts during a roll over accident, human surrogates could exert forces of 40-60 g’s upon a buckle. Peterson then attempted to demonstrate that a buckle would release with forces of only 40-60 g’s, even though prior tests had consistently revealed that buckles would release only when subjected to much higher g forces. In order to demonstrate this, Peterson performed two pendulum tests during which he placed eight pounds of web tension on a seat belt while striking the back of the buckle with a hammer. As a result of these tests, Peterson concluded that, although the buckle was being subjected to peak acceleration forces of up to 200-300 g’s, the buckle was actually releasing at much lower levels of around 40-60 g’s.
To show that Peterson’s tests were unreliable, Ford presented the results of several tests conducted by its experts, the NHTSA and the SAE, in which human surrogates were placed in seat belts during roll over accidents. These tests demonstrated that web tension was always above twenty pounds when the acceleration exceeded 40 g’s. R. at 9:2001, 1923, 1946, 1947, 39:9299. Ford also presented results of tests performed by the SAE in which it subjected a buckle to various web tensions and accelerations to determine the forces needed to release the buckle. These tests demonstrated that even when the buckles were subjected to high accelerations and low web tensions, the two factors had to be present for a sufficient length of time to allow the components within the buckle to move the latch the required distance to release. R. at 9:1940,1947,1987.
Nevertheless, Peterson responded that he believed that it was possible.for a seat belt occupant to exert 40-60 g’s upon a buckle, while, at the same time, exerting little or no tension upon the belt webbing. In support of his theory, Peterson referred to a test in which web tension on a belt went from 0 to 200 pounds almost instantly near the point where the acceleration reached nearly 78 g’s. R. at 40:9321-22. Lytle contends that this demonstrated that the web tension on a belt can remain near zero at acceleration forces greater than 60 g’s. Thus, he concluded that inertial release can occur in a real world accident.
After Peterson’s testimony, the trial court again ruled that his testimony should be excluded, stating as follows:
[t]he forces allowing basis for the opening of the buckle as a result of the pendulum swing, when compared to the forces involved upon'the seat belt buckle when used [by] a human being in a roll over situation are potentially too different to be compared without further evidence and explanation.... The forces conceivably can arise and take place over different periods of time and can last for different periods of time and in other ways might be sufficiently dissimilar that there is no correlation between one set of forces and the other. The court finds therefore the proffered [473]*473testimony to be scientifically unreliable because the inference which the testimony can support is tenuous and speculative at best.
R. at 13:2927-28. We agree with the trial court.
The record reveals that Peterson has not conducted any tests in which he has placed a human surrogate into a seat belt to determine if web tension can remain as low as eight pounds at the same time that the occupant is exerting forces of 40-60 g’s on the buckle. Rather, Lytle contends that Peterson bases his premise on the results of one of Ford’s tests in which human surrogates occupied the seat, belts during roll over accidents. R. at 40:9321-22. While Lytle contends that the web tension was near zero at the same time that the acceleration reached 78 g’s, the documentation of the test makes it impossible to determine with any amount of certainty, whether the tension remained near zero or whether it climbed to 200 pounds at the instant the acceleration reached 78 g’s. R. at 24:5548-49, 39:9269-70. In fact, during the offer of proof, Peterson admitted that the experts who had conducted the test concluded that the tension on the webbing of the belt was above twenty pounds at the instant that the acceleration exceeded 40 g’s. R. at 39:9298-99. Furthermore, the record reveals that the buckle did not release during Ford’s test. This suggests that even if the web tension was near zero while the acceleration reached 78 g’s, the factors did not remain constant long enough for the belt to inertially release. Therefore, despite Lytle’s contentions, this test does not support Peterson’s premise. Rather, it supports the NHTSA’s, SAE’s and Ford’s conclusion that, due to the variables of web tension and acceleration duration, inertial release does not occur in real world accidents. R. at 9:1933, 1947, 2002.
Thus, although Peterson’s pendulum tests are clearly more complex and reliable than Marcosky’s, they serve only to demonstrate that a buckle will inertially release at accelerations of 40-60 g’s when accompanied by low web tension. Peterson’s tests fail to show, however, that web tension - can remain as low as eight pounds at the same time that a person in a rollover accident exerts 40-60 g forces upon the buckle. Rather, prior roll over tests have shown that web tension was always above twenty pounds when the acceleration exceeded 40 g forces. Thus, Peterson has failed to demonstrate an adequate nexus between his pendulum tests and real world accidents. Moreover, Lytle has failed to present the results of any tests in which a buckle inertially released during a roll over accident while a human surrogate was occupying the belt. In light of these circumstances, we cannot conclude that the trial court erred by determining that Peterson’s testimony was not based on reliable scientific principles.8
C. Expert Qualifications
Finally, Lytle contends that the trial court erred by failing to consider Peterson’s and Mareosky’s exceptional credentials in determining the reliability of the scientific principles underlying their testimony regarding inertial release. Specifically, Lytle contends that his experts’ credentials were sufficient to demonstrate the reliability of their scientific principles.
As stated above, the scientific principles upon which an expert bases his testimony must be reliable in order for the expert’s testimony to be admissible. Evid. R. 702(b). In determining reliability, there is no specific “test” or set of “prongs” which must be considered. McGrew v. State, 682 N.E.2d at 1292. However, we have previously considered the following factors: 1) whether the technique has been or can be empirically tested; 2) whether the technique has been subjected to peer review and publication; and 3) general acceptance within the relevant scientific community. See Hottinger, 665 N.E.2d at 596. Although Lytle contends that the court should have also considered his [474]*474experts’ credentials in determining reliability of the scientific principles, Lytle has not presented, and we have not found, any Indiana cases in which we have considered an expert’s credentials in- determining reliability. Although we do not believe that it would be improper for a trial court to consider an expert’s exceptional qualifications in determining the reliability of his scientific testimony, we do not believe that such credentials are conclusive.
Here, the record does not reveal whether the trial court considered the experts’ credentials. However, even assuming the court failed to consider them, we cannot conclude that Mareosky’s and Peterson’s credentials were sufficient to demonstrate the reliability of the scientific principles underlying their respective testimony. As a result, the trial court properly excluded Marcosky’s and Peterson’s scientific testimony regarding inertial release and, therefore, properly granted summary judgment on that theory.
However, we do note that Lytle will be permitted to proceed to trial on his inadvertent release theory. As we stated above, Lytle did not abandon this theory during the hearing on Ford’s motion in limine. Although we also concluded that the trial court properly excluded Mareosky’s testimony on this theory, the record reveals that, in support of his motion in opposition to summary judgment, Lytle submitted Peterson’s affidavit in which he stated that Kyong’s seat belt could have inadvertently released. Because the court did not exclude this testimony, we remand with instructions to proceed on Ly-tle’s theory of inadvertent release.9
CONCLUSION
In summary, we find that the trial court did not exclude all evidence of Kyong’s seat belt use. Additionally, we conclude that the court properly excluded Peterson’s and Mar-cosky’s testimony regarding inertial release and, therefore, the trial court properly granted summary judgment on this theory. However, we conclude that the trial court erred by determining that Lytle had abandoned his theory regarding inadvertent release. Further, although the trial court properly excluded Marcosky’s testimony regarding inadvertent release, the court did not exclude Peterson’s testimony on this theory. Therefore, we remand with instructions that the court proceed on Lytle’s theory of inadvertent release.
Judgment affirmed in part, reversed in part and remanded for proceedings not inconsistent with this opinion.
NAJAM, J., concurs.
RILEY, J., concurring in part and dissenting in part, with opinion.