Chiropractic researchers Robert Mootz and Reed Phillips suggest that, in chiropractic's early years, influences from both straight and mixer concepts were incorporated into its construct. They conclude that chiropractic has both materialistic qualities that lend themselves to scientific investigation and vitalistic qualities that do not.

With relatively little federal funding, academic research in chiropractic has only recently become established in the USA. In 1994 and 1995, half of all grant funding to chiropractic researchers was from the US Health Resources and Services Administration (7 grants totaling $2.3 million). The Foundation for Chiropractic Education and Research (11 grants, $881,000) and the Consortium for Chiropractic Research (4 grants, $519,000) accounted for most of the rest. By 1997, there were 14 peer-reviewed chiropractic journals in English that encouraged the publication of chiropractic research, including The Journal of Manipulative and Physiological Therapeutics (JMPT), Topics in Clinical Chiropractic, and the Journal of Chiropractic Humanities. However, of these, only JMPT is included in Index Medicus. Research into chiropractic, whether from Universities or chiropractic colleges, is however often published in many other scientific journals.

While there is still debate about the effectiveness of chiropractic for the many conditions in which it is applied, chiropractic seems to be most effective for acute low back pain and tension headaches. When testing the efficacy of health treatments, double blind studies are generally considered acceptable scientific rigor. These are designed so that neither the patient nor the doctor knows whether they are using the actual treatment or a placebo (or "sham") treatment. However, chiropractic treatment involves a manipulation; no "sham" procedure can be devised easily for this, and even if the patient is unaware whether the treatment is a real or sham procedure, the doctor cannot be unaware. Thus there may be "observer bias" - the tendency to see what you expect to see, and the potential for the patient to wish to report benefits to "please" the doctor. This problem is not confined to chiropractic - many medical treatments are not amenable to double-blind placebo-controlled trials, indeed this is true for all surgical procedures. It is also a problem in evaluating treatments; even when there are objective outcome measures, the placebo effect can be very substantial. Thus, DCs have historically relied mostly on their own clinical experience and the shared experience of their colleagues, as reported in case studies, to direct their treatment methods. In this, they are not different to the practice in much of conventional medicine.

Sociologist Leslie Biggs interviewed 600 Canadian DCs in 1997: while 86% felt that chiropractic methods needed to be validated, 74% did not believe that controlled clinical trials were the best way to evaluate chiropractic. Moreover, 68% believed that "most diseases are caused by spinal malalignment", although only 30% agreed that "subluxation was the cause of many diseases".

Even when a valid mechanism of action is not determined, it is generally thought sufficient to present evidence showing benefit for the claims made. There is wide agreement that, where applicable, an evidence based medicine framework should be used to assess health outcomes, and that systematic reviews with strict protocols are important for objectively evaluating treatments. Where evidence from such reviews is lacking, this does not necessarily mean that the treatment is ineffective, only that the case for a benefit of treatment may not have been rigorously established.

A 2005 editorial in JMPT, "The Cochrane Collaboration: is it relevant for doctors of chiropractic?" proposed that involvement in Cochrane collaboration would be a way for chiropractic to gain greater acceptance within medicine. The collaboration has 11,500 contributors from more than 90 countries organized in 50 review groups. For chiropractic, relevant review groups include the Back Group; the Bone, Joint, and Muscle Trauma Group; the Musculoskeletal Group; and the Neuromuscular Disease Group. The editorial states that, for example, "a chiropractor may provide conservative care supported by a Cochrane review to a patient with carpal tunnel syndrome. If the patient's symptoms become progressive, the doctor may consider referring the patient for surgery using a recent Cochrane review that examined new surgical techniques compared with traditional open surgery..."

The Cochrane Collaboration did not find enough evidence to support or refute the claim that manual therapy (including, but not limited to, chiropractic) is beneficial for asthma. Carpal tunnel syndrome trials have not shown benefit from diuretics, non-steroidal anti-inflammatory drugs, magnets, laser acupuncture, exercise or chiropractic and there is not enough evidence to show the effects of spinal manipulation (including, but not limited to, chiropractic) for painful menstrual periods. Bandolier found limited evidence that spinal manipulative therapy (including, but not limited to, chiropractic) might reduce the frequency and intensity of migraine attacks, but the evidence that spinal manipulation is better than amitriptyline, or adds to the effects of amitriptyline, is insubstantial for the treatment of migraine, although "spinal manipulative therapy might be worth trying for some patients with migraine or tension headaches."

According to Bandolier, a systematic review of a small, poor quality set of trials, provided no convincing evidence for long-term benefits of chiropractic interventions for acute or chronic low back pain, despite some positive overall findings but there might be some short-term pain relief, especially in patients with acute pain. However, the BMJ noted in a study on long-term low-back problems "...improvement in all patients at three years was about 29% more in those treated by chiropractors than in those treated by the hospitals. The beneficial effect of chiropractic on pain was particularly clear." A 1994 study by the U.S. Agency for Health Care Policy and Research (AHCPR) and the U.S. Department of Health and Human Services endorses spinal manipulation for acute low back pain in adults in its Clinical Practice Guideline.

The first significant recognition of the appropriateness of spinal manipulation for low back pain was performed by the RAND Corporation. This meta-analysis concluded that some forms of spinal manipulation were successful in treating certain types of lower back pain. Some chiropractors claimed these results as proof of chiropractic hypotheses, but RAND's studies were about spinal manipulation, not chiropractic specifically, and dealt with appropriateness, which is a measure of net benefit and harms; the efficacy of chiropractic and other treatments were not explicitly compared. In 1993, Dr Shekelle rebuked some DCs for their exaggerated claims: ...we have become aware of numerous instances where our results have been seriously misrepresented by chiropractors writing for their local paper or writing letters to the editor....

There is conflict in the results of chiropractic research. For instance, many DCs claim to treat infantile colic. According to a 1999 survey, 46% of chiropractors in Ontario treated children for colic. In 1999 a Danish randomized controlled clinical trial with a blinded observer suggested that there is evidence that spinal manipulation might help infantile colic. However, in 2001, a Norwegian blinded study concluded that chiropractic spinal manipulation was no more effective than placebo for treating infantile colic.

In 1997, historian Joseph Keating Jr described chiropractic as a "science, antiscience and pseudoscience", and said "Although available scientific data support chiropractic's principle intervention method (the manipulation of patients with lower back pain), the doubting, skeptical attitudes of science do not predominate in chiropractic education or among practitioners". He argued that chiropractic's culture has nurtured antiscientific attitudes and activities, and that "a combination of uncritical rationalism and uncritical empiricism has been bolstered by the proliferation of pseudoscience journals of chiropractic wherein poor quality research and exuberant over-interpretation of results masquerade as science and provide false confidence about the value of various chiropractic techniques". However, in 1998, after reviewing the articles published in the JMPT from 1989-1996, he concluded,

"substantial increases in scholarly activities within the chiropractic profession are suggested by the growth in scholarly products published in the discipline's most distinguished periodical (JMPT). Increases in controlled outcome studies, collaboration among chiropractic institutions, contributions from nonchiropractors, contributions from nonchiropractic institutions and funding for research suggest a degree of professional maturation and growing interest in the content of the discipline."