A hypothesis is a testable prediction about a relationship between variables, stated precisely enough that data can contradict it. Vague hypotheses cannot be tested, and hypotheses that merely restate the research question add nothing to a thesis.
Enter what you want to test and this tool drafts matched null and alternative pairs for each main relationship type, naming the statistical test that goes with each — along with the reporting conventions that cause trouble in results chapters.
The null hypothesis states there is no difference, relationship or effect; the alternative states there is one. You test the null, because it is the specific claim data can contradict — a test calculates how likely your data would be if the null were true.
This governs how you write your conclusion. You either reject the null or fail to reject it. You never prove the alternative and never prove the null: failing to reject may only mean your sample was too small to detect an effect that exists.
A difference between two groups needs an independent-samples t-test; three or more, a one-way ANOVA. A relationship between two continuous variables needs Pearson correlation, or Spearman where data is ordinal or badly skewed. An effect of one variable on another needs regression.
An association between two categorical variables needs a chi-square test of independence. Choosing the test before collecting data is not premature — it dictates what you measure and at what level. Discovering afterwards that your data cannot support the intended test is common and entirely avoidable.
Only if your study is quantitative and tests a specific prediction. Exploratory and qualitative studies use research questions, and forcing hypotheses onto interview-based work is a category error examiners notice at once.
Descriptive quantitative work sits between: if you are establishing how common something is, a research question serves better. Hypotheses belong where you compare, correlate or test an effect.
A two-tailed hypothesis states there is a difference without specifying direction, and is the default. A one-tailed hypothesis predicts direction and concentrates statistical power on that side.
That makes one-tailed tests easier to reach significance with, which is exactly why they require justification from prior literature rather than convenience. Choosing one-tailed after seeing which way your data leans is a serious error, and a detectable one.
Fix your significance level in advance, conventionally 0.05. Report the test statistic, degrees of freedom, p-value and an effect size. A p-value tells you whether an effect is distinguishable from zero; it says nothing about magnitude or importance.
With a large enough sample, trivial differences reach statistical significance. Effect size — Cohen's d, eta squared, r — tells the reader whether a finding matters in practice, and examiners increasingly expect it as standard rather than as an optional extra.
The null states there is no difference, relationship or effect; the alternative states there is one. You test the null, because it is the claim data can contradict.
No. Qualitative and exploratory studies use research questions. Hypotheses belong to quantitative work testing specific predictions.
No. You reject or fail to reject the null. Failing to reject is not evidence the null is true — it may just mean insufficient power.
Usually one per objective. More than four or five in a Master's thesis normally means the scope is too broad.
0.05 is conventional in social science. Choose it before analysis, never after seeing your results.