Question

Health research methods

• Why do you think so many hospital administrators have difficulty understanding the statistical methods used in healthcare research?

Provide examples of concepts (such as a confidence interval) and practical skills you have learned that could be used to explain the information to healthcare leaders who might not have an MHA or training in healthcare research. •

Jacobsen, K. H. (2017). Introduction to health research methods (2nd ed.). Burlington, MA: Jones & Bartlett Learning.

Answer 1

Statistics in health research

Def :Statistics is a branch of applied mathematics that deals with collecting, organizing, and interpreting data using well-defined procedures.

Data: is all the information needed and collected for the research purpose.

Methods of Statistics :

                                    1.Descriptive statistics

                                    2. Inferential Statistics

Descriptive statistics : Described and the data is arranged with thehelp of scales as required,

The various scales are:

• Nominal scale
• Ordinal scale
• Interval scale &
• Ratio scale

consists simple discription of data methods .They are

1. Meassures of central tendency. Eg; mean, median and mode
2. Messures of variance or dispertion.Eg; range, standared deviation,mean deviation, quartile deviation ,etc.
3. Messures of relationship .Eg; Correlation Coefficiant, Regressioon analysis

Inferential Statistics: After organising date , then inferential statistics is used to analyse the data and systamatically draw the results.

The hypothesis formulated in the beginning of the study is tested using various inferential methods

Depending upon the parameters used , devided into

1. Parametric tests
2. Non parametric tests

parametric tests, which are characterized by three attributes: (1) they involve the estimation of a parameter; (2) they require measurements on at least an interval scale; and (3) they involve several assumptions, such as the assumption that the variables are normally distributed in the population.

The various parametric tests are

1. T test, paired and unpaired
2. Z test
3. ANOVA-, ANCOVA,MANOVA

Nonparametric tests, by contrast, do not estimate parameters. They are usually applied when data have been measured on a nominal or ordinal scale. Nonparametric methods involve less restrictive assumptions about the shape of the variables’ distribution than do parametric tests. For this reason,

nonparametric tests are sometimes called distribution-free statistics.

The various non parametric tests are:

1. Chi square test
2. Maan wheitny u test
3. Rank order test ,ect

The common terms used in statistics are;

1. CI , confidence interval: is a type of interval estimate (of a population parameter) that is computed from the observed data. The confidence level is the frequency (i.e., the proportion) of possible confidence intervals that contain the true value of their corresponding parameter.
2. Type one and type two errors : Type I error by rejecting the null hypothesis when it is, in fact, true. For instance, if we concluded that the experimental treatment was more effective than the control condition in alleviating patients’ anxiety, when in fact observed differences in anxiety scores resulted from sampling fluctuations, we would be making a Type I error.

Type II error Conversely, if we concluded that group differences in anxiety scores resulted by chance, when in fact the intervention did reduce anxiety, we would be committing a Type II error by accepting a false.

1. Level of Significance Researchers do not know when an error in statistical decision making has been made. The validity of a null hypothesis could be ascertained only by collecting data from the population, in which case there would be no need for statistical inference Researchers do, however, control the risk of a Type I error by selecting a level of significance, which signifies the probability of rejecting a true null hypothesis.

The two most frequently used significance levels (referred to as alpha or _) are .05 and .01.

Eg: With a .05 significance level, we are accepting the risk that out of 100 samples drawn from a population, a true null hypothesis would be rejected only 5 times. With a .01 significance level, the risk of a Type I error is lower: in only 1 sample out of 100 would we erroneously reject the null hypothesis.

The minimum acceptable level for _ usually is .05. A stricter level (e.g., .01 or .001) may be needed when the decision has important consequences