Documente Academic
Documente Profesional
Documente Cultură
Link to full text Link to external site, this link will open in a new window
Turn on hit highlighting for speaking browsers by selecting the Enter button
Hide highlighting
Abstract
Translate Abstract
There is widespread evidence of the extensive use of statistical methods in medical
research. Just the same, standards are generally low and a growing body of literature
points to statistical errors in most medical journals. However, there is no
comprehensive study contrasting the top medical journals of basic and clinical
science for recent practice in their use of statistics. All original research articles in
Volume 10, Numbers 1-6 of Nature Medicine (Nat Med) and Volume 350, Numbers
1-26 of The New England Journal of Medicine (NEJM) were screened for their
statistical content. Types, frequencies, and complexity of applied statistical methods
were systematically recorded. A 46-item checklist was used to evaluate statistical
quality for a subgroup of papers. 94.5 percent (95% CI 87.6-98.2) of NEJM articles
and 82.4 percent (95% CI 65.5-93.2) of Nat Med articles contained
inferential statistics. NEJM papers were significantly more likely to use advanced
statistical methods (p < 0.0001). Statistical errors were identified in a considerable
proportion of articles, although not always serious in nature.
Documentation of applied statistical methods was generally poor and insufficient,
particularly in Nat Med. Compared to 1983, a vast increase in usage and
complexity of statistical methods could be observed for NEJM papers. This does not
necessarily hold true for Nat Med papers, as the results of the study indicate that
basic science sticks with basic analysis. As statistical errors seem to remain common
in medical literature, closer attention to statistical methodology should be seriously
considered to raise standards.
Another issue to be considered in deciding the part insights courses ought to take
in a business degree project is the subject of which topical territories ought to be
taught. A distinguishing proof of the basic points that ought to be incorporated into
the undergrad measurements educational modules will decide the quantity of
courses required and maybe give understanding into the office and/or school in
which insights ought to be taught.
Further the teaching content, method and data analysis of statistics in a business
degree program are discussed.
In spite of the fact that the present study gives data with respect to the cutting edge
in the educating of measurements, it has certain restrictions. In the first place, in
spite of the fact that the study is exploratory in nature, the examination just
approached respondents for their view of the significance of different subjects and
whether they incorporated those themes in their courses. These discernments and
their recollections could be obfuscated by some type of attractive quality
inclination, which could impact their reactions with respect to which points are
incorporated and which are most vital in the course. In spite of the fact that this is a
typical deficiency of study research, the discoveries do give knowledge into the
impression of significance and consideration from the suppliers' points of view. A
second restriction is that the concentrate just evaluated the impression of AACSB
educators, who, one could contend, speak to just a minority of the considerable
number of schools and colleges occupied with showing business measurements. By
and by, on the grounds that these instructors may be portrayed as fitting in with the
more elite class of advanced education, they ought to be the ones most mindful of
applicable changes and necessities in business training.
Future examination ought to address these confinements. The specimen and
information gathering configuration could likewise be changed both to extend the
example and to build the reaction rate. Future studies may incorporate the view of
business college graduates; truth be told, graduates could be overviewed in a
longitudinal study to survey their utilization of measurements and their suppositions
as their vocations progress.
The results of the study indicate that 83 (72.8%) of the responding universities
provided undergraduate students with two statistics courses and 79 (69.3%)
provided a graduate level statistics course. These findings support the notion that the
majority of the respondents at AACSB universities believe that statistics is an
important tool course that should be a part of their core business offerings.
Further, of the 17 statistical topics examined, there seemed to be a rather clear
separation of topics by level of the course. In Table l, we show the topic areas
taught in each statistics course.
The most commonly taught topics in the introductory course (Statistics 1) included
descriptive statistics (97.2%), probability distribution (95.5%), estimation and
hypothesis testing (90.9%), and tables and charts (90.0%). Other topics generally
covered (over 50%) in the Statistics 1 course were statistical software, exploratory
data analysis, and linear regression and correlation. Though other topics were
occasionally noted, they were reported as receiving much less coverage in the
first statistics course.
Respondents indicated that the second course (Statistics 2) tended to concentrate on
higher level, or multivariate, techniques such as multiple regressions (89.2%), linear
regression and correlation (83. No), statistical software (80.7%),
analysis of variance (73.5%), and the use of technology applications in problem
solving. Time series and estimation and hypothesis testing were the only other
topics covered by at least 50% of the responding universities. Some of the topics
reportedly covered in the Statistics 1 course also were given coverage, in some
cases, but the overall percentage of respondents noting those topics was relatively
small. Graduate programs tended to cover a wide range of topics, with 10 of our
list of 17 covered by more than 50% of the responding universities, likely because
graduate level business statistics courses are frequently part of the master's of
business administration course work and contain a substantial review component.
The second area covered in this study, the level of importance
that statistics professors believed each topic has to the business community, was
measured through a 5-point Likert scale ranging from I (not important) to 5 (very
important). The results of this portion of the questionnaire appear in Table 2.
Of the 17 topics examined, eight received mean ratings of 4.00 or better, indicating
that they were viewed as being of relatively high importance to
the business community. These topics were, in order of importance,
descriptive statistics, linear regression and correlation, tables and charts, multiple
regression, estimation and hypothesis testing, statistical software, quality and
productivity, and time series. The other nine topics were rated less important to
the business community and ranged from exploratory data analysis (3.87) to factor
analysis (2.32). Perhaps as a result of this perceived importance, five of the eight
topics most highly rated (i.e., descriptive statistics. linear regression, tables and
charts, estimation and hypothesis testing, and statistical software) are generally
covered in the Statistics 1 course. Two other topics (multiple regression and time
series) were reported to be covered in the Statistics 2 course. Only quality and
productivity, which received a 4.16 rating, was included by less than 50% of the
respondents in either the Statistics I or Statistics 2 course.
Discussion
An interpretation of the findings may lead to some interesting implications. First,
when subjects that professors regarded as most important are compared with the
topics most frequently included in their courses, some differences are obvious. For
example, probability distribution was identified as being included in almost
96% of the introductory statistics courses, yet it was only tenth in its rating on
importance (X (over scored) = 3.76). Further, linear regression and correlation was
identified as the second most important topic (X (over scored) = 4.48), and multiple
regression was identified as the fourth most important (X = 4.26); yet these two
topics were ranked seventh (53.6%) and tenth (20.0%) in terms of their
inclusion in the introductory statistics course. These deviations may be interpreted
to indicate conflicting conclusions. The first conclusion, which seems most
plausible to us, is that two statistics courses may be necessary. This conclusion is
based on the premise that the first course must provide the basics of statistics,
regardless of their importance in future applications. It seems myopic to suggest that
because the topic of probability distribution received only a
rank of tenth in importance it should not be included in the
introductory statistics course. Without it, students would have little
comprehension of the foundation for much of their future statistical analyses. A
similar conclusion could be drawn regarding exploratory data analysis, as its
exclusion could arguably undermine the basis for much of the logic and theory that
follows. Yet, as any professor recognizes, a limited resource in the
introductory statistics course is almost always time; there are too many critical
subjects to discuss and too little time
Conversely, one could reasonably argue that too much time is devoted to the
theory of statistics and too little to its practice. Though we may disagree with this
particular conclusion, its logic may be deserving of some discussion. The findings
indicate that the most important topics seem to be related to the practice of
statistics (i.e., descriptive statistics, linear regression, tables and charts, multiple
regression, etc.). These topics could be emphasized to the exclusion of more
theoretical topics (i.e., probability distribution, estimation and hypothesis testing,
nonparametric statistics), and the student may still be able to obtain valuable
information that could be applied in the business community.
As stated, we believe that these findings provide a persuasive argument for a
second statistics course. A failure to provide the students with the basic foundation