State in your own words what i

Type I error -Error that occurs when the researcher concludes that the sampletested are from different population (a significant differenceexist between groups) when, in fact, the samples are from the samepopulation (no significant difference exist between groups); thenull hypothesis is rejected when it is true.

Type II error –Error that occurs when the researcher concludes that no significantdifference exists between the samples examined when, in fact, adifference exists; the null hypothesis is regarded as true when itis false.

Hypothesis testing is the sheetanchor of empirical research and in the rapidly emerging practiceof evidence-based medicine.The empirical approach to researchcannot eliminate uncertainty completely. At the best, it canquantify uncertainty. This uncertainty can be of 2 types: Type Ierror (falsely rejecting a null hypothesis) and type II error(falsely accepting a null hypothesis). The acceptable magnitudes oftype I and type II errors are set in advance and are important forsample size calculations.

• Validity refers towhether in an experiment a tool actually measures what it claims toand whether the results could actually be generalised to the widerworld.
• There are two main types ofvalidity – internal and external.
• Internal validityrelates to whether changes made to a dependent variable actuallychange and influence the independent variable – in other wordswhether the tool measures what it is meant to. If there is highinternal validity, you could expect the dependent variable tomanipulate the independent variable, whereas if there was lowinternal validity it would be suspected that confounding variableswere playing a part in effecting the independent variable.
• Internal validity can beimproved in a few simple ways. In order to ensure aninvestigating is measuring what it is meant to, investigators canuse single and double-blind techniques. A single-blind study iswhere the participant does not know the condition they are in, anddouble-blind is where neither the participant nor the experimenterknows what the groups represent. This method ensures there islittle demand characteristics, such as trying to behave a certainway because they think that is what is expected of them, and alsodecreases experimenter effects, as they can not even accidentlyhave a bias to a certain group. Another process that is effectivewhen a repeated measure has been used is counterbalancing andrandom allocation, as they remove order effects, such as boredom,fatigue.